research paper on genetics

• Search Menu
• Advance Articles
• Perspectives
• Knowledgebase and Database Resources
• Nobel Laureates Collection

China Virtual Outreach Webinar

Neurogenetics, fungal genetics and genomics.

• Multiparental Populations
• Genomic Prediction
• Plant Genetics and Genomics

Genetic Models of Rare Diseases

• Genomic Data Analyses In Biobanks
• Why Publish
• Author Guidelines
• Submission Site
• Open Access Options
• Full Data Policy
• Self-Archiving Policy
• About Genetics
• About Genetics Society of America
• Editorial Board
• Early Career Reviewers
• Guidelines for Reviewers
• Advertising & Corporate Services
• Journals on Oxford Academic
• Books on Oxford Academic

Editor-in-Chief

Howard Lipshitz

Executive Editor

Tracey DePellegrin

Managing Editor

Ruth Isaacson

Scientific Editor and Program Manager

Opportunities and challenges for genomic data analyses in biobanks: a call for papers.

The GSA Journals are calling for submissions of papers on biobank-scale genomic data analyses. The closing date for submissions is May 31 2024.

Why publish with GENETICS?

Why publish in genetics.

Learn more about why GENETICS is the perfect home for your research, and submit today to join our celebrated author community.

Why publish?

Series and Collections accepting papers

Submit your work to one of GSA’s ongoing series and collections.

Currently accepting submissions

Meet the Editorial Board

See who handles papers for GENETICS by topic.

Editorial board

Re-watch the recent China Virtual Outreach Webinar where you will learn more about publishing your work in the journal.

Watch the webinar

Latest articles

Series & collections.

Genes and variants of interest in rare diseases often benefit from modelling in cellular assays or genetic models to aid in understanding molecular and cellular mechanisms of disfunction. Model organisms are useful for the discovery of new genetic diseases and key to understanding variant effects, and modelling a disease gene in a genetic model means that researchers can perform an in-depth exploration of gene or variant function. The GSA Journals are pleased to publish a series highlighting ongoing advances in rare disease discovery and mechanisms by presenting key research findings and new discoveries.

Plant Genetics and Genomics 

Plant science has generated many discoveries and advances in genetics and genomics research. These contributions reflect the ingenuity and rigor of the plant science community, as well as the rich diversity of plants and their biology. To showcase this critical work, GENETICS and G3: Genes|Genomes|Genetics has launched the Plant Genetics and Genomics series with a collection of fourteen research articles and an accompanying editorial.

Neurogenetics lies at the intersection of Neuroscience and Genetics, where genetic approaches are applied to the study of nervous system development, function, and plasticity. Overseen by Series Editors Oliver Hobert, Cecilia Moens, and Kate O’Connor Giles, this new series aims to make the GSA Journals a home for cutting-edge, robust research in neurogenetics.

The fungal kingdom is remarkable in its breadth and depth of impact on global health, agriculture, biodiversity, ecology, manufacturing, and biomedical research. Overseen by editors Leah Cowen and Joseph Heitman, this series aims to report and thereby further stimulate advances in genetics and genomics across a diversity of fungal species.

FlyBook from GENETICS is a comprehensive compendium of review articles presenting the current state of knowledge in  Drosophila  research.

Browse FlyBook

WormBook from GENETICS features a comprehensive compendium of review articles presenting the current state of knowledge in  C. elegans  research. WormBook articles will span the breadth of the biology, genetics, genomics, and evolutionary biology of  C. elegans .

Browse WormBook

The YeastBook series from GENETICS features a comprehensive compendium of reviews that presents the current state of knowledge of the molecular biology, cellular biology, and genetics of the yeast  Saccharomyces cerevisiae .

Browse YeastBook

More from GSA

G3: Genes|Genomes|Genetics

G3, a Genetics Society of America journal, provides a forum for the publication of high-quality foundational research-particularly research that generates useful genetic and genomic information, as well as genome reports, mutant screens, and advances in methods and technology.

Find out more

GSA members of all career stages receive member benefits including access to professional development programs, discounted meeting registration, and eligibility for travel awards. Members also receive a personal subscription to GENETICS, as well as discounted publication fees in both GSA journals.

Conferences

GSA conferences have long served as community hubs for researchers focused on particular organisms or topics. GSA also hosts The Allied Genetics Conference (TAGC) , a unique meeting that brings together multiple research communities for collaboration and synthesis.

Attend a conference

Career Development

GSA professional development programs provide rich opportunities for scientists to gain skills,  experience, mentors, and networks. Our initiatives and resources range from peer review training to inclusive public engagement, newsletters, webinars, a job board, leadership programs, and much more.

Browse Opportunities

Email alerts

Register to receive email alerts as soon as new content from  GENETICS is published online.

Recommend to your library

Fill out our simple online form to recommend GENETICS to your library. Recommend now

Author resources

Learn about how to submit your article, our publishing process, and tips on how to promote your article.

Related Titles

• Recommend to Your Librarian
• Advertising and Corporate Services
• Journals Career Network

Affiliations

• Online ISSN 1943-2631
• Copyright © 2024 Genetics Society of America
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

The genetic basis of disease

Affiliations.

• 1 School of Medicine, Dentistry and Nursing, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow G12 8QQ, U.K.
• 2 School of Medicine, Dentistry and Nursing, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow G12 8QQ, U.K. [email protected].
• 3 School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow G12 8QQ, U.K.
• PMID: 30509934
• PMCID: PMC6279436
• DOI: 10.1042/EBC20170053

Genetics plays a role, to a greater or lesser extent, in all diseases. Variations in our DNA and differences in how that DNA functions (alone or in combinations), alongside the environment (which encompasses lifestyle), contribute to disease processes. This review explores the genetic basis of human disease, including single gene disorders, chromosomal imbalances, epigenetics, cancer and complex disorders, and considers how our understanding and technological advances can be applied to provision of appropriate diagnosis, management and therapy for patients.

Keywords: cancer; genetics; genomics; molecular basis of health and disease.

© 2018 The Author(s).

Publication types

• Chromosome Aberrations
• DNA / genetics
• Disease Models, Animal
• Epigenesis, Genetic
• Genetic Diseases, Inborn / diagnosis
• Genetic Diseases, Inborn / genetics*
• Genetic Diseases, Inborn / therapy
• Genetic Variation
• Genome, Human
• Neoplasms / genetics
• Polymerase Chain Reaction

Advertisement

A review on genetic algorithm: past, present, and future

• Published: 31 October 2020
• Volume 80 , pages 8091–8126, ( 2021 )

Cite this article

• Sourabh Katoch 1 ,
• Sumit Singh Chauhan 1 &
• Vijay Kumar   ORCID: orcid.org/0000-0002-3460-6989 1  

147k Accesses

1670 Citations

14 Altmetric

Explore all metrics

In this paper, the analysis of recent advances in genetic algorithms is discussed. The genetic algorithms of great interest in research community are selected for analysis. This review will help the new and demanding researchers to provide the wider vision of genetic algorithms. The well-known algorithms and their implementation are presented with their pros and cons. The genetic operators and their usages are discussed with the aim of facilitating new researchers. The different research domains involved in genetic algorithms are covered. The future research directions in the area of genetic operators, fitness function and hybrid algorithms are discussed. This structured review will be helpful for research and graduate teaching.

Similar content being viewed by others

Genetic algorithms: theory, genetic operators, solutions, and applications

Bushra Alhijawi & Arafat Awajan

Genetic Algorithms

Structure and Operation of a Basic Genetic Algorithm

Avoid common mistakes on your manuscript.

1 Introduction

In the recent years, metaheuristic algorithms are used to solve real-life complex problems arising from different fields such as economics, engineering, politics, management, and engineering [ 113 ]. Intensification and diversification are the key elements of metaheuristic algorithm. The proper balance between these elements are required to solve the real-life problem in an effective manner. Most of metaheuristic algorithms are inspired from biological evolution process, swarm behavior, and physics’ law [ 17 ]. These algorithms are broadly classified into two categories namely single solution and population based metaheuristic algorithm (Fig.  1 ). Single-solution based metaheuristic algorithms utilize single candidate solution and improve this solution by using local search. However, the solution obtained from single-solution based metaheuristics may stuck in local optima [ 112 ]. The well-known single-solution based metaheuristics are simulated annealing, tabu search (TS), microcanonical annealing (MA), and guided local search (GLS). Population-based metaheuristics utilizes multiple candidate solutions during the search process. These metaheuristics maintain the diversity in population and avoid the solutions are being stuck in local optima. Some of well-known population-based metaheuristic algorithms are genetic algorithm (GA) [ 135 ], particle swarm optimization (PSO) [ 101 ], ant colony optimization (ACO) [ 47 ], spotted hyena optimizer (SHO) [ 41 ], emperor penguin optimizer (EPO) [ 42 ], and seagull optimization (SOA) [ 43 ].

Classification of metaheuristic Algorithms

Among the metaheuristic algorithms, Genetic algorithm (GA) is a well-known algorithm, which is inspired from biological evolution process [ 136 ]. GA mimics the Darwinian theory of survival of fittest in nature. GA was proposed by J.H. Holland in 1992. The basic elements of GA are chromosome representation, fitness selection, and biological-inspired operators. Holland also introduced a novel element namely, Inversion that is generally used in implementations of GA [ 77 ]. Typically, the chromosomes take the binary string format. In chromosomes, each locus (specific position on chromosome) has two possible alleles (variant forms of genes) - 0 and 1. Chromosomes are considered as points in the solution space. These are processed using genetic operators by iteratively replacing its population. The fitness function is used to assign a value for all the chromosomes in the population [ 136 ]. The biological-inspired operators are selection, mutation, and crossover. In selection, the chromosomes are selected on the basis of its fitness value for further processing. In crossover operator, a random locus is chosen and it changes the subsequences between chromosomes to create off-springs. In mutation, some bits of the chromosomes will be randomly flipped on the basis of probability [ 77 , 135 , 136 ]. The further development of GA based on operators, representation, and fitness has diminished. Therefore, these elements of GA are focused in this paper.

The main contribution of this paper are as follows:

The general framework of GA and hybrid GA are elaborated with mathematical formulation.

The various types of genetic operators are discussed with their pros and cons.

The variants of GA with their pros and cons are discussed.

The applicability of GA in multimedia fields is discussed.

The main aim of this paper is two folds. First, it presents the variants of GA and their applicability in various fields. Second, it broadens the area of possible users in various fields. The various types of crossover, mutation, selection, and encoding techniques are discussed. The single-objective, multi-objective, parallel, and hybrid GAs are deliberated with their advantages and disadvantages. The multimedia applications of GAs are elaborated.

The remainder of this paper is organized as follows: Section 2 presents the methodology used to carry out the research. The classical genetic algorithm and genetic operators are discussed in Section 3 . The variants of genetic algorithm with pros and cons are presented in Section 4 . Section 5 describes the applications of genetic algorithm. Section 6 presents the challenges and future research directions. The concluding remarks are drawn in Section 7 .

2 Research methodology

PRISMA’s guidelines were used to conduct the review of GA [ 138 ]. A detailed search has been done on Google scholar and PubMed for identification of research papers related to GA. The important research works found during the manual search were also added in this paper. During search, some keywords such as “Genetic Algorithm” or “Application of GA” or “operators of GA” or “representation of GA” or “variants of GA” were used. The selection and rejection of explored research papers are based on the principles, which is mentioned in Table 1 .

Total 27,64,792 research papers were explored on Google Scholar, PubMed and manual search. The research work related to genetic algorithm for multimedia applications were also included. During the screening of research papers, all the duplicate papers and papers published before 2007 were discarded. 4340 research papers were selected based on 2007 and duplicate entries. Thereafter, 4050 research papers were eliminated based on titles. 220 research papers were eliminated after reading of abstract. 70 research papers were left after third round of screening. 40 more research papers were discarded after full paper reading and facts found in the papers. After the fourth round of screening, final 30 research papers are selected for review.

Based on the relevance and quality of research, 30 papers were selected for evaluation. The relevance of research is decided through some criteria, which is mentioned in Table 1 . The selected research papers comprise of genetic algorithm for multimedia applications, advancement of their genetic operators, and hybridization of genetic algorithm with other well-established metaheuristic algorithms. The pros and cons of genetic operators are shown in preceding section.

3 Background

In this section, the basic structure of GA and its genetic operators are discussed with pros and cons.

3.1 Classical GA

Genetic algorithm (GA) is an optimization algorithm that is inspired from the natural selection. It is a population based search algorithm, which utilizes the concept of survival of fittest [ 135 ]. The new populations are produced by iterative use of genetic operators on individuals present in the population. The chromosome representation, selection, crossover, mutation, and fitness function computation are the key elements of GA. The procedure of GA is as follows. A population ( Y ) of n chromosomes are initialized randomly. The fitness of each chromosome in Y is computed. Two chromosomes say C1 and C2 are selected from the population Y according to the fitness value. The single-point crossover operator with crossover probability (C p ) is applied on C1 and C2 to produce an offspring say O . Thereafter, uniform mutation operator is applied on produced offspring ( O ) with mutation probability (M p ) to generate O′ . The new offspring O′ is placed in new population. The selection, crossover, and mutation operations will be repeated on current population until the new population is complete. The mathematical analysis of GA is as follows [ 126 ]:

GA dynamically change the search process through the probabilities of crossover and mutation and reached to optimal solution. GA can modify the encoded genes. GA can evaluate multiple individuals and produce multiple optimal solutions. Hence, GA has better global search capability. The offspring produced from crossover of parent chromosomes is probable to abolish the admirable genetic schemas parent chromosomes and crossover formula is defined as [ 126 ]:

where g is the number of generations, and G is the total number of evolutionary generation set by population. It is observed from Eq.( 1 ) that R is dynamically changed and increase with increase in number of evolutionary generation. In initial stage of GA, the similarity between individuals is very low. The value of R should be low to ensure that the new population will not destroy the excellent genetic schema of individuals. At the end of evolution, the similarity between individuals is very high as well as the value of R should be high.

According to Schema theorem, the original schema has to be replaced with modified schema. To maintain the diversity in population, the new schema keep the initial population during the early stage of evolution. At the end of evolution, the appropriate schema will be produced to prevent any distortion of excellent genetic schema [ 65 , 75 ]. Algorithm 1 shows the pseudocode of classical genetic algorithm.

Algorithm 1: Classical Genetic Algorithm (GA)

3.2 Genetic operators

GAs used a variety of operators during the search process. These operators are encoding schemes, crossover, mutation, and selection. Figure 2 depicts the operators used in GAs.

Operators used in GA

3.2.1 Encoding schemes

For most of the computational problems, the encoding scheme (i.e., to convert in particular form) plays an important role. The given information has to be encoded in a particular bit string [ 121 , 183 ]. The encoding schemes are differentiated according to the problem domain. The well-known encoding schemes are binary, octal, hexadecimal, permutation, value-based, and tree.

Binary encoding is the commonly used encoding scheme. Each gene or chromosome is represented as a string of 1 or 0 [ 187 ]. In binary encoding, each bit represents the characteristics of the solution. It provides faster implementation of crossover and mutation operators. However, it requires extra effort to convert into binary form and accuracy of algorithm depends upon the binary conversion. The bit stream is changed according the problem. Binary encoding scheme is not appropriate for some engineering design problems due to epistasis and natural representation.

In octal encoding scheme, the gene or chromosome is represented in the form of octal numbers (0–7). In hexadecimal encoding scheme, the gene or chromosome is represented in the form of hexadecimal numbers (0–9, A-F) [ 111 , 125 , 187 ]. The permutation encoding scheme is generally used in ordering problems. In this encoding scheme, the gene or chromosome is represented by the string of numbers that represents the position in a sequence. In value encoding scheme, the gene or chromosome is represented using string of some values. These values can be real, integer number, or character [ 57 ]. This encoding scheme can be helpful in solving the problems in which more complicated values are used. As binary encoding may fail in such problems. It is mainly used in neural networks for finding the optimal weights.

In tree encoding, the gene or chromosome is represented by a tree of functions or commands. These functions and commands can be related to any programming language. This is very much similar to the representation of repression in tree format [ 88 ]. This type of encoding is generally used in evolving programs or expressions. Table 2 shows the comparison of different encoding schemes of GA.

3.2.2 Selection techniques

Selection is an important step in genetic algorithms that determines whether the particular string will participate in the reproduction process or not. The selection step is sometimes also known as the reproduction operator [ 57 , 88 ]. The convergence rate of GA depends upon the selection pressure. The well-known selection techniques are roulette wheel, rank, tournament, boltzmann, and stochastic universal sampling.

Roulette wheel selection maps all the possible strings onto a wheel with a portion of the wheel allocated to them according to their fitness value. This wheel is then rotated randomly to select specific solutions that will participate in formation of the next generation [ 88 ]. However, it suffers from many problems such as errors introduced by its stochastic nature. De Jong and Brindle modified the roulette wheel selection method to remove errors by introducing the concept of determinism in selection procedure. Rank selection is the modified form of Roulette wheel selection. It utilizes the ranks instead of fitness value. Ranks are given to them according to their fitness value so that each individual gets a chance of getting selected according to their ranks. Rank selection method reduces the chances of prematurely converging the solution to a local minima [ 88 ].

Tournament selection technique was first proposed by Brindle in 1983. The individuals are selected according to their fitness values from a stochastic roulette wheel in pairs. After selection, the individuals with higher fitness value are added to the pool of next generation [ 88 ]. In this method of selection, each individual is compared with all n-1 other individuals if it reaches the final population of solutions [ 88 ]. Stochastic universal sampling (SUS) is an extension to the existing roulette wheel selection method. It uses a random starting point in the list of individuals from a generation and selects the new individual at evenly spaced intervals [ 3 ]. It gives equal chance to all the individuals in getting selected for participating in crossover for the next generation. Although in case of Travelling Salesman Problem, SUS performs well but as the problem size increases, the traditional Roulette wheel selection performs relatively well [ 180 ].

Boltzmann selection is based on entropy and sampling methods, which are used in Monte Carlo Simulation. It helps in solving the problem of premature convergence [ 118 ]. The probability is very high for selecting the best string, while it executes in very less time. However, there is a possibility of information loss. It can be managed through elitism [ 175 ]. Elitism selection was proposed by K. D. Jong (1975) for improving the performance of Roulette wheel selection. It ensures the elitist individual in a generation is always propagated to the next generation. If the individual having the highest fitness value is not present in the next generation after normal selection procedure, then the elitist one is also included in the next generation automatically [ 88 ]. The comparison of above-mentioned selection techniques are depicted in Table 3 .

3.2.3 Crossover operators

Crossover operators are used to generate the offspring by combining the genetic information of two or more parents. The well-known crossover operators are single-point, two-point, k-point, uniform, partially matched, order, precedence preserving crossover, shuffle, reduced surrogate and cycle.

In a single point crossover, a random crossover point is selected. The genetic information of two parents which is beyond that point will be swapped with each other [ 190 ]. Figure 3 shows the genetic information after swapping. It replaced the tail array bits of both the parents to get the new offspring.

Swapping genetic information after a crossover point

In a two point and k-point crossover, two or more random crossover points are selected and the genetic information of parents will be swapped as per the segments that have been created [ 190 ]. Figure 4 shows the swapping of genetic information between crossover points. The middle segment of the parents is replaced to generate the new offspring.

Swapping genetic information between crossover points

In a uniform crossover, parent cannot be decomposed into segments. The parent can be treated as each gene separately. We randomly decide whether we need to swap the gene with the same location of another chromosome [ 190 ]. Figure 5 depicts the swapping of individuals under uniform crossover operation.

Swapping individual genes

Partially matched crossover (PMX) is the most frequently used crossover operator. It is an operator that performs better than most of the other crossover operators. The partially matched (mapped) crossover was proposed by D. Goldberg and R. Lingle [ 66 ]. Two parents are choose for mating. One parent donates some part of genetic material and the corresponding part of other parent participates in the child. Once this process is completed, the left out alleles are copied from the second parent [ 83 ]. Figure 6 depicts the example of PMX.

Partially matched crossover (PMX) [ 117 ]

Order crossover (OX) was proposed by Davis in 1985. OX copies one (or more) parts of parent to the offspring from the selected cut-points and fills the remaining space with values other than the ones included in the copied section. The variants of OX are proposed by different researchers for different type of problems. OX is useful for ordering problems [ 166 ]. However, it is found that OX is less efficient in case of Travelling Salesman Problem [ 140 ]. Precedence preserving crossover (PPX) preserves the ordering of individual solutions as present in the parent of offspring before the application of crossover. The offspring is initialized to a string of random 1’s and 0’s that decides whether the individuals from both parents are to be selected or not. In [ 169 ], authors proposed a modified version of PPX for multi-objective scheduling problems.

Shuffle crossover was proposed by Eshelman et al. [ 20 ] to reduce the bias introduced by other crossover techniques. It shuffles the values of an individual solution before the crossover and unshuffles them after crossover operation is performed so that the crossover point does not introduce any bias in crossover. However, the utilization of this crossover is very limited in the recent years. Reduced surrogate crossover (RCX) reduces the unnecessary crossovers if the parents have the same gene sequence for solution representations [ 20 , 139 ]. RCX is based on the assumption that GA produces better individuals if the parents are sufficiently diverse in their genetic composition. However, RCX cannot produce better individuals for those parents that have same composition. Cycle crossover was proposed by Oliver [ 140 ]. It attempts to generate an offspring using parents where each element occupies the position by referring to the position of their parents [ 140 ]. In the first cycle, it takes some elements from the first parent. In the second cycle, it takes the remaining elements from the second parent as shown in Fig.  7 .

Cycle Crossover (CX) [ 140 ]

Table 4 shows the comparison of crossover techniques. It is observed from Table 4 that single and k-point crossover techniques are easy to implement. Uniform crossover is suitable for large subsets. Order and cycle crossovers provide better exploration than the other crossover techniques. Partially matched crossover provides better exploration. The performance of partially matched crossover is better than the other crossover techniques. Reduced surrogate and cycle crossovers suffer from premature convergence.

3.2.4 Mutation operators

Mutation is an operator that maintains the genetic diversity from one population to the next population. The well-known mutation operators are displacement, simple inversion, and scramble mutation. Displacement mutation (DM) operator displaces a substring of a given individual solution within itself. The place is randomly chosen from the given substring for displacement such that the resulting solution is valid as well as a random displacement mutation. There are variants of DM are exchange mutation and insertion mutation. In Exchange mutation and insertion mutation operators, a part of an individual solution is either exchanged with another part or inserted in another location, respectively [ 88 ].

The simple inversion mutation operator (SIM) reverses the substring between any two specified locations in an individual solution. SIM is an inversion operator that reverses the randomly selected string and places it at a random location [ 88 ]. The scramble mutation (SM) operator places the elements in a specified range of the individual solution in a random order and checks whether the fitness value of the recently generated solution is improved or not [ 88 ]. Table 5 shows the comparison of different mutation techniques.

Table 6 shows the best combination of encoding scheme, mutation, and crossover techniques. It is observed from Table 6 that uniform and single-point crossovers can be used with most of encoding and mutation operators. Partially matched crossover is used with inversion mutation and permutation encoding scheme provides the optimal solution.

4 Variants of GA

Various variants of GA’s have been proposed by researchers. The variants of GA are broadly classified into five main categories namely, real and binary coded, multiobjective, parallel, chaotic, and hybrid GAs. The pros and cons of these algorithms with their application has been discussed in the preceding subsections.

4.1 Real and binary coded GAs

Based on the representation of chromosomes, GAs are categorized in two classes, namely binary and real coded GAs.

4.1.1 Binary coded GAs

The binary representation was used to encode GA and known as binary GA. The genetic operators were also modified to carry out the search process. Payne and Glen [ 153 ] developed a binary GA to identify the similarity among molecules. They used binary representation for position of molecule and their conformations. However, this method has high computational complexity. Longyan et al. [ 203 ] investigated three different method for wind farm design using binary GA (BGA). Their method produced better fitness value and farm efficiency. Shukla et al. [ 185 ] utilized BGA for feature subset selection. They used mutual information maximization concept for selecting the significant features. BGAs suffer from Hamming cliffs, uneven schema, and difficulty in achieving precision [ 116 , 199 ].

4.1.2 Real-coded GAs

Real-coded GAs (RGAs) have been widely used in various real-life applications. The representation of chromosomes is closely associated with real-life problems. The main advantages of RGAs are robust, efficient, and accurate. However, RGAs suffer from premature convergence. Researchers are working on RGAs to improve their performance. Most of RGAs are developed by modifying the crossover, mutation and selection operators.

Crossover operators

The searching capability of crossover operators are not satisfactory for continuous search space. The developments in crossover operators have been done to enhance their performance in real environment. Wright [ 210 ] presented a heuristics crossover that was applied on parents to produce off-spring. Michalewicz [ 135 ] proposed arithmetical crossover operators for RGAs. Deb and Agrawal [ 34 ] developed a real-coded crossover operator, which is based on characteristics of single-point crossover in BGA. The developed crossover operator named as simulated binary crossover (SBX). SBX is able to overcome the Hamming cliff, precision, and fixed mapping problem. The performance of SBX is not satisfactory in two-variable blocked function. Eshelman et al. [ 53 ] utilized the schemata concept to design the blend crossover for RGAs. The unimodal normal distribution crossover operator (UNDX) was developed by Ono et al. [ 144 ]. They used ellipsoidal probability distribution to generate the offspring. Kita et al. [ 106 ] presented a multi-parent UNDX (MP-UNDX), which is the extension of [ 144 ]. However, the performance of RGA with MP-UNDX is much similar to UNDX. Deep and Thakur [ 39 ] presented a Laplace crossover for RGAs, which is based on Laplacian distribution. Chuang et al. [ 27 ] developed a direction based crossover to further explore the all possible search directions. However, the search directions are limited. The heuristic normal distribution crossover operator was developed by Wang et al. [ 207 ]. It generates the cross-generated offspring for better search operation. However, the better individuals are not considered in this approach. Subbaraj et al. [ 192 ] proposed Taguchi self-adaptive RCGA. They used Taguchi method and simulated binary crossover to exploit the capable offspring.

Mutation operators

Mutation operators generate diversity in the population. The two main challenges have to tackle during the application of mutation. First, the probability of mutation operator that was applied on population. Second, the outlier produced in chromosome after mutation process. Michalewicz [ 135 ] presented uniform and non-uniform mutation operators for RGAs. Michalewicz and Schoenauer [ 136 ] developed a special case of uniform mutation. They developed boundary mutation. Deep and Thakur [ 38 ] presented a novel mutation operator based on power law and named as power mutation. Das and Pratihar [ 30 ] presented direction-based exponential mutation operator. They used direction information of variables. Tang and Tseng [ 196 ] presented a novel mutation operator for enhancing the performance of RCGA. Their approach was fast and reliable. However, it stuck in local optima for some applications. Deb et al. [ 35 ] developed polynomial mutation that was used in RCGA. It provides better exploration. However, the convergence speed is slow and stuck in local optima. Lucasius et al. [ 129 ] proposed a real-coded genetic algorithm (RCGA). It is simple and easy to implement. However, it suffers from local optima problem. Wang et al. [ 205 ] developed multi-offspring GA and investigated their performance over single point crossover. Wang et al. [ 206 ] stated the theoretical basis of multi-offspring GA. The performance of this method is better than non-multi-offspring GA. Pattanaik et al. [ 152 ] presented an improvement in the RCGA. Their method has better convergence speed and quality of solution. Wang et al. [ 208 ] proposed multi-offspring RCGA with direction based crossover for solving constrained problems.

Table 7 shows the mathematical formulation of genetic operators in RGAs.

4.2 Multiobjective GAs

Multiobjective GA (MOGA) is the modified version of simple GA. MOGA differ from GA in terms of fitness function assignment. The remaining steps are similar to GA. The main motive of multiobjective GA is to generate the optimal Pareto Front in the objective space in such a way that no further enhancement in any fitness function without disturbing the other fitness functions [ 123 ]. Convergence, diversity, and coverage are main goal of multiobjective GAs. The multiobjective GAs are broadly categorized into two categories namely, Pareto-based, and decomposition-based multiobjective GAs [ 52 ]. These techniques are discussed in the preceding subsections.

4.2.1 Pareto-based multi-objective GA

The concept of Pareto dominance was introduced in multiobjective GAs. Fonseca and Fleming [ 56 ] developed first multiobjective GA (MOGA). The niche and decision maker concepts were proposed to tackle the multimodal problems. However, MOGA suffers from parameter tuning problem and degree of selection pressure. Horn et al. [ 80 ] proposed a niched Pareto genetic algorithm (NPGA) that utilized the concept of tournament selection and Pareto dominance. Srinivas and Deb [ 191 ] developed a non-dominated sorting genetic algorithm (NSGA). However, it suffers from lack of elitism, need of sharing parameter, and high computation complexity. To alleviate these problems, Deb et al. [ 36 ] developed a fast elitist non-dominated sorting genetic algorithm (NSGA-II). The performance of NSGA-II may be deteriorated for many objective problems. NSGA-II was unable to maintain the diversity in Pareto-front. To alleviate this problem, Luo et al. [ 130 ] introduced a dynamic crowding distance in NSGA-II. Coello and Pulido [ 28 ] developed a multiobjective micro GA. They used an archive for storing the non-dominated solutions. The performance of Pareto-based approaches may be deteriorated in many objective problems [ 52 ].

4.2.2 Decomposition-based multiobjective GA

Decomposition-based MOGAs decompose the given problem into multiple subproblems. These subproblems are solved simultaneously and exchange the solutions among neighboring subproblems [ 52 ]. Ishibuchi and Murata [ 84 ] developed a multiobjective genetic local search (MOGLS). In MOGLS, the random weights were used to select the parents and local search for their offspring. They used generation replacement and roulette wheel selection method. Jaszkiewicz [ 86 ] modified the MOGLS by utilizing different selection mechanisms for parents. Murata and Gen [ 141 ] proposed a cellular genetic algorithm for multiobjective optimization (C-MOGA) that was an extension of MOGA. They added cellular structure in MOGA. In C-MOGA, the selection operator was performed on the neighboring of each cell. C-MOGA was further extended by introducing an immigration procedure and known as CI-MOGA. Alves and Almeida [ 11 ] developed a multiobjective Tchebycheffs-based genetic algorithm (MOTGA) that ensures convergence and diversity. Tchebycheff scalar function was used to generate non-dominated solution set. Patel et al. [ 151 ] proposed a decomposition based MOGA (D-MOGA). They integrated opposition based learning in D-MOGA for weight vector generation. D-MOGA is able to maintain the balance between diversity of solutions and exploration of search space.

4.3 Parallel GAs

The motivation behind the parallel GAs is to improve the computational time and quality of solutions through distributed individuals. Parallel GAs are categorized into three broad categories such as master-slave parallel GAs, fine grained parallel GAs, and multi-population coarse grained parallel Gas [ 70 ]. In master-slave parallel GA, the computation of fitness functions is distributed over the several processors. In fine grained GA, parallel computers are used to solve the real-life problems. The genetic operators are bounded to their neighborhood. However, the interaction is allowed among the individuals. In coarse grained GA, the exchange of individuals among sub-populations is performed. The control parameters are also transferred during migration. The main challenges in parallel GAs are to maximize memory bandwidth and arrange threads for utilizing the power of GPUs [ 23 ]. Table 8 shows the comparative analysis of parallel GAs in terms of hardware and software. The well-known parallel GAs are studied in the preceding subsections.

4.3.1 Master slave parallel GA

The large number of processors are utilized in master-slave parallel GA (MS-PGA) as compared to other approaches. The computation of fitness functions may be increased by increasing the number of processors. Hong et al. [ 79 ] used MS-PGA for solving data mining problems. Fuzzy rules are used with parallel GA. The evaluation of fitness function was performed on slave machines. However, it suffers from high computational time. Sahingzo [ 174 ] implemented MS-PGA for UAV path finding problem. The genetic operators were executed on processors. They used multicore CPU with four cores. Selection and fitness evaluation was done on slave machines. MS-PGA was applied on traffic assignment problem in [ 127 ]. They used thirty processors to solve this problem at National University of Singapore. Yang et al. [ 213 ] developed a web-based parallel GA. They implemented the master slave version of NSGA-II in distributed environment. However, the system is complex in nature.

4.3.2 Fine grained parallel GA

In last few decades, researchers are working on migration policies of fine grained parallel GA (FG-PGA). Porta et al. [ 161 ] utilized clock-time for migration frequency, which is independent of generations. They used non-uniform structure and static configuration. The best solution was selected for migration and worst solution was replaced with migrant solution. Kurdi [ 115 ] used adaptive migration frequency. The migration procedure starts until there is no change in the obtained solutions after ten successive generations. The non-uniform and dynamic structure was used. In [ 209 ], local best solutions were synchronized and formed a global best solutions. The global best solutions were transferred to all processors for father execution. The migration frequency depends upon the number of generation. They used uniform structure with fixed configuration. Zhang et al. [ 220 ] used parallel GA to solve the set cover problem of wireless networks. They used divide-and-conquer strategy to decompose the population into sub-populations. Thereafter, the genetic operators were applied on local solutions and Kuhn-Munkres was used to merge the local solutions.

4.3.3 Coarse grained parallel GA

Pinel et al. [ 158 ] proposed a GraphCell. The population was initialized with random values and one solution was initialized with Min-min heuristic technique. 448 processors were used to implement the proposed approach. However, coarse grained parallel GAs are less used due to complex in nature. The hybrid parallel GAs are widely used in various applications. Shayeghi et al. [ 182 ] proposed a pool-based Birmingham cluster GA. Master node was responsible for managing global population. Slave node selected the solutions from global population and executed it. 240 processors are used for computation. Roberge et al. [ 170 ] used hybrid approach to optimize switching angle of inverters. They used four different strategies for fitness function computation. Nowadays, GPU, cloud, and grid are most popular hardware for parallel GAs [ 198 ].

4.4 Chaotic GAs

The main drawback of GAs is premature convergence. The chaotic systems are incorporated into GAs to alleviate this problem. The diversity of chaos genetic algorithm removes premature convergence. Crossover and mutation operators can be replaced with chaotic maps. Tiong et al. [ 197 ] integrated the chaotic maps into GA for further improvement in accuracy. They used six different chaotic maps. The performance of Logistic, Henon and Ikeda chaotic GA performed better than the classical GA. However, these techniques suffer from high computational complexity. Ebrahimzadeh and Jampour [ 48 ] used Lorenz chaotic for genetic operators of GA to eliminate the local optima problem. However, the proposed approach was unable to find relationship between entropy and chaotic map. Javidi and Hosseinpourfard [ 87 ] utilized two chaotic maps namely logistic map and tent map for generating chaotic values instead of random selection of initial population. The proposed chaotic GA performs better than the GA. However, this method suffers from high computational complexity. Fuertes et al. [ 60 ] integrated the entropy into chaotic GA. The control parameters are modified through chaotic maps. They investigated the relationship between entropy and performance optimization.

Chaotic systems have also used in multiobjective and hybrid GAs. Abo-Elnaga and Nasr [ 5 ] integrated chaotic system into modified GA for solving Bi-level programming problems. Chaotic helps the proposed algorithm to alleviate local optima and enhance the convergence. Tahir et al. [ 193 ] presented a binary chaotic GA for feature selection in healthcare. The chaotic maps were used to initialize the population and modified reproduction operators were applied on population. Xu et al. [ 115 ] proposed a chaotic hybrid immune GA for spectrum allocation. The proposed approach utilizes the advantages of both chaotic and immune operator. However, this method suffers from parameter initialization problem.

4.5 Hybrid GAs

Genetic Algorithms can be easily hybridized with other optimization methods for improving their performance such as image denoising methods, chemical reaction optimization, and many more. The main advantages of hybridized GA with other methods are better solution quality, better efficiency, guarantee of feasible solutions, and optimized control parameters [ 51 ]. It is observed from literature that the sampling capability of GAs is greatly affected from population size. To resolve this problem, local search algorithms such as memetic algorithm, Baldwinian, Lamarckian, and local search have been integrated with GAs. This integration provides proper balance between intensification and diversification. Another problem in GA is parameter setting. Finding appropriate control parameters is a tedious task. The other metaheuristic techniques can be used with GA to resolve this problem. Hybrid GAs have been used to solve the issues mentioned in the preceding subsections [ 29 , 137 , 186 ].

4.5.1 Enhance search capability

GAs have been integrated with local search algorithms to reduce the genetic drift. The explicit refinement operator was introduced in local search for producing better solutions. El-Mihoub et al. [ 54 ] established the effect of probability of local search on the population size of GA. Espinoza et al. [ 50 ] investigated the effect of local search for reducing the population size of GA. Different search algorithms have been integrated with GAs for solving real-life applications.

4.5.2 Generate feasible solutions

In complex and high-dimensional problems, the genetic operators of GA generate infeasible solutions. PMX crossover generates the infeasible solutions for order-based problems. The distance preserving crossover operator was developed to generate feasible solutions for travelling salesman problem [ 58 ]. The gene pooling operator instead of crossover was used to generate feasible solution for data clustering [ 19 ]. Konak and Smith [ 108 ] integrated a cut-saturation algorithm with GA for designing the communication networks. They used uniform crossover to produce feasible solutions.

4.5.3 Replacement of genetic operators

There is a possibility to replace the genetic operators which are mentioned in Section 3.2 with other search techniques. Leng [ 122 ] developed a guided GA that utilizes the penalties from guided local search. These penalties were used in fitness function to improve the performance of GA. Headar and Fukushima [ 74 ] used simplex crossover instead of standard crossover. The standard mutation operator was replaced with simulated annealing in [ 195 ]. The basic concepts of quantum computing are used to improve the performance of GAs. The heuristic crossover and hill-climbing operators can be integrated into GA for solving three-matching problem.

4.5.4 Optimize control parameters

The control parameters of GA play a crucial role in maintaining the balance between intensification and diversification. Fuzzy logic has an ability to estimate the appropriate control parameters of GA [ 167 ]. Beside this, GA can be used to optimize the control parameters of other techniques. GAs have been used to optimize the learning rate, weights, and topology of neutral networks [ 21 ]. GAs can be used to estimate the optimal value of fuzzy membership in controller. It was also used to optimize the control parameters of ACO, PSO, and other metaheuristic techniques [ 156 ]. The comparative analysis of well-known GAs are mentioned in Table 9 .

5 Applications

Genetic Algorithms have been applied in various NP-hard problems with high accuracy rates. There are a few application areas in which GAs have been successfully applied.

5.1 Operation management

GA is an efficient metaheuristic for solving operation management (OM) problems such as facility layout problem (FLP), supply network design, scheduling, forecasting, and inventory control.

5.1.1 Facility layout

Datta et al. [ 32 ] utilized GA for solving single row facility layout problem (SRFLP). For SRFLP, the modified crossover and mutation operators of GA produce valid solutions. They applied GA to large sized problems that consists of 60–80 instances. However, it suffers from parameter dependency problem. Sadrzadeh [ 173 ] proposed GA for multi-line FLP have multi products. The facilities were clustered using mutation and heuristic operators. The total cost obtained from the proposed GA was decreased by 7.2% as compared to the other algorithms. Wu et al. [ 211 ] implemented hierarchical GA to find out the layout of cellular manufacturing system. However, the performance of GA is greatly affected from the genetic operators. Aiello et al. [ 7 ] proposed MOGA for FLP. They used MOGA on the layout of twenty different departments. Palomo-Romero et al. [ 148 ] proposed an island model GA to solve the FLP. The proposed technique maintains the population diversity and generates better solutions than the existing techniques. However, this technique suffers from improper migration strategy that can be utilized for improving the population. GA and its variants has been successfully applied on FLP [ 103 , 119 , 133 , 201 ].

5.1.2 Scheduling

GA shows the superior performance for solving the scheduling problems such as job-shop scheduling (JSS), integrated process planning and scheduling (IPPS), etc. [ 119 ]. To improve the performance in the above-mentioned areas of scheduling, researchers developed various genetic representation [ 12 , 159 , 215 ], genetic operators, and hybridized GA with other methods [ 2 , 67 , 147 , 219 ].

5.1.3 Inventory control

Besides the scheduling, inventory control plays an important role in OM. Backordering and lost sales are two main approaches for inventory control [ 119 ]. Hiassat et al. [ 76 ] utilized the location-inventory model to find out the number and location of warehouses. Various design constraints have been added in the objective functions of GA and its variants for solving inventory control problem [].

5.1.4 Forecasting and network design

Forecasting is an important component for OM. Researchers are working on forecasting of financial trading, logistics demand, and tourist arrivals. GA has been hybridized with support vector regression, fuzzy set, and neural network (NN) to improve their forecasting capability [ 22 , 78 , 89 , 178 , 214 ]. Supply network design greatly affect the operations planning and scheduling. Most of the research articles are focused on capacity constraints of facilities [ 45 , 184 ]. Multi-product multi-period problems increases the complexity of supply networks. To resolve the above-mentioned problem, GA has been hybridized with other techniques [ 6 , 45 , 55 , 188 , 189 ]. Multi-objective GAs are also used to optimize the cost, profit, carbon emissions, etc. [ 184 , 189 ].

5.2 Multimedia

GAs have been applied in various fields of multimedia. Some of well-known multimedia fields are encryption, image processing, video processing, medical imaging, and gaming.

5.2.1 Information security

Due to development in multimedia applications, images, videos and audios are transferred from one place to another over Internet. It has been found in literature that the images are more error prone during the transmission. Therefore, image protection techniques such as encryption, watermarking and cryptography are required. The classical image encryption techniques require the input parameters for encryption. The wrong selection of input parameters will generate inadequate encryption results. GA and its variants have been used to select the appropriate control parameters. Kaur and Kumar [ 96 ] developed a multi-objective genetic algorithm to optimize the control parameters of chaotic map. The secret key was generated using beta chaotic map. The generated key was use to encrypt the image. Parallel GAs were also used to encrypt the image [ 97 ].

5.2.2 Image processing

The main image processing tasks are preprocessing, segmentation, object detection, denoising, and recognition. Image segmentation is an important step to solve the image processing problems. Decomposing/partitioning an image requires high computational time. To resolve this problem, GA is used due to their better search capability [ 26 , 102 ]. Enhancement is a technique to improve the quality and contrast of an image. The better image quality is required to analyze the given image. GAs have been used to enhance natural contrast and magnify image [ 40 , 64 , 99 ]. Some researchers are working on hybridization of rough set with adaptive genetic algorithm to merge the noise and color attributes. GAs have been used to remove the noise from the given image. GA can be hybridized with fuzzy logic to denoise the noisy image. GA based restoration technique can be used to remove haze, fog and smog from the given image [ 8 , 110 , 146 , 200 ]. Object detection and recognition is a challenging issue in real-world problem. Gaussian mixture model provides better performance during detection and recognition process. The control parameters are optimized through GA [ 93 ].

5.2.3 Video processing

Video segmentation has been widely used in pattern recognition, and computer vision. There are some critical issues that are associated with video segmentation. These are distinguishing object from the background and determine accurate boundaries. GA can be used to resolve these issues [ 9 , 105 ]. GAs have been implemented for gesture recognition successfully by Chao el al. [ 81 ] used GA for gesture recognition. They applied GAs and found an accuracy of 95% in robot vision. Kaluri and Reddy [ 91 ] proposed an adaptive genetic algorithm based method along with fuzzy classifiers for sign gesture recognition. They reported an improved recognition rate of 85% as compared to the existing method that provides 79% accuracy. Beside the gesture recognition, face recognition play an important role in criminal identification, unmanned vehicles, surveillance, and robots. GA is able to tackle the occlusion, orientations, expressions, pose, and lighting condition [ 69 , 95 , 109 ].

5.2.4 Medical imaging

Genetic algorithms have been applied in medical imaging such as edge detection in MRI and pulmonary nodules detection in CT scan images [ 100 , 179 ]. In [ 120 ], authors used a template matching technique with GA for detecting nodules in CT images. Kavitha and Chellamuthu [ 179 ] used GA based region growing method for detecting the brain tumor. GAs have been applied on medical prediction problems captured from pathological subjects. Sari and Tuna [ 176 ] used GA used to solve issues arises in biomechanics. It is used to predict pathologies during examination. Ghosh and Bhattachrya [ 62 ] implemented sequential GA with cellular automata for modelling the coronavirus disease 19 (COVID-19) data. GAs can be applied in parallel mode to find rules in biological datasets [ 31 ]. The authors proposed a parallel GA that runs by dividing the process into small sub-generations and evaluating the fitness of each individual solution in parallel. Genetic algorithms are used in medicine and other related fields. Koh et al. [ 61 ] proposed a genetic algorithm based method for evaluation of adverse effects of a given drug.

5.2.5 Precision agriculture

GAs have been applied on various problems that are related to precision agriculture. The main issues are crop yield, weed detection, and improvement in farming equipment. Pachepsky and Acock [ 145 ] implemented GA to analyze the water capacity in soil using remote sensing images. The crop yield can be predicted through the capacity of water present in soil. The weed identification was done through GA in [ 142 ]. They used aerial image for classification of plants. In [ 124 ], color image segmentation was used to discriminate the weed and plant. Peerlink et al. [ 154 ] determined the appropriate rate of fertilizer for various portions of agriculture field. They GA for determining the nitrogen in wheat field. The energy requirements in water irrigation systems can be optimized by viewing it as a multi-objective optimization problem. The amount of irrigation required and thus power requirements change continuously in a SMART farm. Therefore, GA can be applied in irrigation systems to reduce the power requirements [ 33 ].

5.2.6 Gaming

GAs have been successfully used in games such as gomoku. In [ 202 ], the authors shown that the GA based approach finds the solution having the highest fitness than the normal tree based methods. However, in real-time strategy based games, GA based solutions become less practical to implement [ 82 ]. GAs have been implemented for path planning problems considering the environment constraints as well as avoiding the obstacles to reach the given destination. Burchardt and Salomon [ 18 ] described an implementation for path planning for soccer games. GA can encode the path planning problems via the coordinate points of a two-dimensional playing field, hence resulting in a variable length solution. The fitness function in path planning considers length of path as well as the collision avoiding terms for soccer players.

5.3 Wireless networking

Due to adaptive, scalable, and easy implementation of GA, it has been used to solve the various issues of wireless networking. The main issues of wireless networking are routing, quality of service, load balancing, localization, bandwidth allocation and channel assignment [ 128 , 134 ]. GA has been hybridized with other metaheuristics for solving the routing problems. Hybrid GA not only producing the efficient routes among pair of nodes, but also used for load balancing [ 24 , 212 ].

5.3.1 Load balancing

Nowadays, multimedia applications require Quality-of-Service (QoS) demand for delay and bandwidth. Various researchers are working on GAs for QoS based solutions.GA produces optimal solutions for complex networks [ 49 ]. Roy et al. [ 172 ] proposed a multi-objective GA for multicast QoS routing problem. GA was used with ACO and other search algorithms for finding optimal routes with desired QoS metrics. Load balancing is another issue in wireless networks. Scully and Brown [ 177 ] used MicroGAs and MacroGAs to distribute the load among various components of networks. He et al. [ 73 ] implemented GA to determine the balance load in wireless sensor networks. Cheng et al. [ 25 ] utilized distributed GA with multi-population scheme for load balancing. They used load balancing metric as a fitness function in GA.

5.3.2 Localization

The process of determining the location of wireless nodes is called as localization. It plays an important role in disaster management and military services. Yun et al. [ 216 ] used GA with fuzzy logic to find out the weights, which are assigned according to the signal strength. Zhang et al. [ 218 ] hybridized GA with simulated annealing (SA) to determine the position of wireless nodes. SA is used as local search to eliminate the premature convergence.

5.3.3 Bandwidth and channel allocation

The appropriate bandwidth allocation is a complex task. GAs and its variants have been developed to solve the bandwidth allocation problem [ 92 , 94 , 107 ]. GAs were used to investigate the allocation of bandwidth with QoS constraints. The fitness function of GAs may consists of resource utilization, bandwidth distribution, and computation time [ 168 ]. The channel allocation is an important issue in wireless networks. The main objective of channel allocation is to simultaneously optimize the number of channels and reuse of allocated frequency. Friend et al. [ 59 ] used distributed island GA to resolve the channel allocation problem in cognitive radio networks. Zhenhua et al. [ 221 ] implemented a modified immune GA for channel assignment. They used different encoding scheme and immune operators. Pinagapany and Kulkarni [ 157 ] developed a parallel GA to solve both static and dynamic channel allocation problem. They used decimal encoding scheme. Table 10 summarizes the applications of GA and its variants.

6 Challenges and future possibilities

In this section, the main challenges faced during the implementation of GAs are discussed followed by the possible research directions.

6.1 Challenges

Despite the several advantages, there are some challenges that need to be resolved for future advancements and further evolution of genetic algorithms. Some major challenges are given below:

6.1.1 Selection of initial population

Initial population is always considered as an important factor for the performance of genetic algorithms. The size of population also affects the quality of solution [ 160 ]. The researchers argue that if a large population is considered, then the algorithm takes more computation time. However, the small population may lead to poor solution [ 155 ]. Therefore, finding the appropriate population size is always a challenging issue. Harik and Lobo [ 71 ] investigated the population using self-adaption method. They used two approaches such as (1) use of self-adaption prior to execution of algorithm, in which the size of population remains the same and (2) in which the self-adaption used during the algorithm execution where the population size is affected by fitness function.

6.1.2 Premature convergence

Premature convergence is a common issue for GA. It can lead to the loss of alleles that makes it difficult to identify a gene [ 15 ]. Premature convergence states that the result will be suboptimal if the optimization problem coincides too early. To avoid this issue, some researchers suggested that the diversity should be used. The selection pressure should be used to increase the diversity. Selection pressure is a degree which favors the better individuals in the initial population of GA’s. If selection pressure (SP1) is greater than some selection pressure (SP2), then population using SP1 should be larger than the population using SP2. The higher selection pressure can decrease the population diversity that may lead to premature convergence [ 71 ].

Convergence property has to be handled properly so that the algorithm finds global optimal solution instead of local optimal solution (see Fig. 8 ). If the optimal solution lies in the vicinity of an infeasible solution, then the global nature of GA can be combined with local nature of other algorithms such as Tabu search and local search. The global nature of genetic algorithms and local nature of Tabu search provide the proper balance between intensification and diversification.

Local and global optima [ 149 ]

6.1.3 Selection of efficient fitness functions

Fitness function is the driving force, which plays an important role in selecting the fittest individual in every iteration of an algorithm. If the number of iterations are small, then a costly fitness function can be adjusted. The number of iterations increases may increase the computational cost. The selection of fitness function depends upon the computational cost as well as their suitability. In [ 46 ], the authors used Davies-Bouldin index for classification of documents.

6.1.4 Degree of mutation and crossover

Crossover and mutation operators are the integral part of GAs. If the mutation is not considered during evolution, then there will be no new information available for evolution. If crossover is not considered during evolution, then the algorithm can result in local optima. The degree of these operators greatly affect the performance of GAs [ 72 ]. The proper balance between these operators are required to ensure the global optima. The probabilistic nature cannot determine the exact degree for an effective and optimal solution.

6.1.5 Selection of encoding schemes

GAs require a particular encoding scheme for a specific problem. There is no general methodology for deciding whether the particular encoding scheme is suitable for any type of real-life problem. If there are two different problems, then two different encoding schemes are required. Ronald [ 171 ] suggested that the encoding schemes should be designed to overwhelm the redundant forms. The genetic operators should be implemented in a manner that they are not biased towards the redundant forms.

6.2 Future research directions

GAs have been applied in different fields by modifying the basic structure of GA. The optimality of a solution obtained from GA can be made better by overcoming the current challenges. Some future possibilities for GA are as follows:

There should be some way to choose the appropriate degree of crossover and mutation operators. For example Self-Organizing GA adapt the crossover and mutation operators according to the given problem. It can save computation time that make it faster.

Future work can also be considered for reducing premature convergence problem. Some researchers are working in this direction. However, it is suggested that new methods of crossover and mutation techniques are required to tackle the premature convergence problem.

Genetic algorithms mimic the natural evolution process. There can be a possible scope for simulating the natural evolution process such as the responses of human immune system and the mutations in viruses.

In real-life problems, the mapping from genotype to phenotype is complex. In this situation, the problem has no obvious building blocks or building blocks are not adjacent groups of genes. Hence, there is a possibility to develop novel encoding schemes to different problems that does not exhibit same degree of difficulty.

7 Conclusions

This paper presents the structured and explained view of genetic algorithms. GA and its variants have been discussed with application. Application specific genetic operators are discussed. Some genetic operators are designed for representation. However, they are not applicable to research domains. The role of genetic operators such as crossover, mutation, and selection in alleviating the premature convergence is studied extensively. The applicability of GA and its variants in various research domain has been discussed. Multimedia and wireless network applications were the main attention of this paper. The challenges and issues mentioned in this paper will help the practitioners to carry out their research. There are many advantages of using GAs in other research domains and metaheuristic algorithms.

The intention of this paper is not only provide the source of recent research in GAs, but also provide the information about each component of GA. It will encourage the researchers to understand the fundamentals of GA and use the knowledge in their research problems.

Abbasi M, Rafiee M, Khosravi MR, Jolfaei A, Menon VG, Koushyar JM (2020) An efficient parallel genetic algorithm solution for vehicle routing problem in cloud implementation of the intelligent transportation systems. Journal of cloud Computing 9(6)

Abdelghany A, Abdelghany K, Azadian F (2017) Airline flight schedule planning under competition. Comput Oper Res 87:20–39

MathSciNet   MATH   Google Scholar  

Abdulal W, Ramachandram S (2011) Reliability-aware genetic scheduling algorithm in grid environment. International Conference on Communication Systems and Network Technologies, Katra, Jammu, pp 673–677

Google Scholar  

Abdullah J (2010) Multiobjectives ga-based QoS routing protocol for mobile ad hoc network. Int J Grid Distrib Comput 3(4):57–68

Abo-Elnaga Y, Nasr S (2020) Modified evolutionary algorithm and chaotic search for Bilevel programming problems. Symmetry 12:767

Afrouzy ZA, Nasseri SH, Mahdavi I (2016) A genetic algorithm for supply chain configuration with new product development. Comput Ind Eng 101:440–454

Aiello G, Scalia G (2012) La, Enea M. A multi objective genetic algorithm for the facility layout problem based upon slicing structure encoding Expert Syst Appl 39(12):10352–10358

Alaoui A, Adamou-Mitiche ABH, Mitiche L (2020) Effective hybrid genetic algorithm for removing salt and pepper noise. IET Image Process 14(2):289–296

Alkhafaji BJ, Salih MA, Nabat ZM, Shnain SA (2020) Segmenting video frame images using genetic algorithms. Periodicals of Engineering and Natural Sciences 8(2):1106–1114

Al-Oqaily AT, Shakah G (2018) Solving non-linear optimization problems using parallel genetic algorithm. International Conference on Computer Science and Information Technology (CSIT), Amman, pp. 103–106

Alvesa MJ, Almeidab M (2007) MOTGA: A multiobjective Tchebycheff based genetic algorithm for the multidimensional knapsack problem. Comput Oper Res 34:3458–3470

MathSciNet   Google Scholar  

Arakaki RK, Usberti FL (2018) Hybrid genetic algorithm for the open capacitated arc routing problem. Comput Oper Res 90:221–231

Arkhipov DI, Wu D, Wu T, Regan AC (2020) A parallel genetic algorithm framework for transportation planning and logistics management. IEEE Access 8:106506–106515

Azadeh A, Elahi S, Farahani MH, Nasirian B (2017) A genetic algorithm-Taguchi based approach to inventory routing problem of a single perishable product with transshipment. Comput Ind Eng 104:124–133

Baker JE, Grefenstette J (2014) Proceedings of the first international conference on genetic algorithms and their applications. Taylor and Francis, Hoboken, pp 101–105

Bolboca SD, JAntschi L, Balan MC, Diudea MV, Sestras RE (2010) State of art in genetic algorithms for agricultural systems. Not Bot Hort Agrobot Cluj 38(3):51–63

Bonabeau E, Dorigo M, Theraulaz G (1999) Swarm intelligence: from natural to artificial systems. Oxford University Press, Inc

MATH   Google Scholar  

Burchardt H, Salomon R (2006) Implementation of path planning using genetic algorithms on Mobile robots. IEEE International Conference on Evolutionary Computation, Vancouver, BC, pp 1831–1836

Burdsall B, Giraud-Carrier C (1997) Evolving fuzzy prototypes for efficient data clustering," in second international ICSC symposium on fuzzy logic and applications. Zurich, Switzerland, pp. 217-223.

Burkowski FJ (1999) Shuffle crossover and mutual information. Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406), Washington, DC, USA, 1999, pp. 1574–1580

Chaiyaratana N, Zalzala AM (2000) "Hybridisation of neural networks and a genetic algorithm for friction compensation," in the 2000 congress on evolutionary computation, vol 1. San Diego, USA, pp 22–29

Chen R, Liang C-Y, Hong W-C, Gu D-X (2015) Forecasting holiday daily tourist flow based on seasonal support vector regression with adaptive genetic algorithm. Appl Soft Comput 26:434–443

J.R. Cheng and M. Gen (2020) Parallel genetic algorithms with GPU computing. Impact on Intelligent Logistics and Manufacturing.

Cheng H, Yang S (2010) Multi-population genetic algorithms with immigrants scheme for dynamic shortest path routing problems in mobile ad hoc networks. Applications of evolutionary computation. Springer, In, pp 562–571

Cheng H, Yang S, Cao J (2013) Dynamic genetic algorithms for the dynamic load balanced clustering problem in mobile ad hoc net-works. Expert Syst Appl 40(4):1381–1392

Chouhan SS, Kaul A, Singh UP (2018) Soft computing approaches for image segmentation: a survey. Multimed Tools Appl 77(21):28483–28537

Chuang YC, Chen CT, Hwang C (2016) A simple and efficient real-coded genetic algorithm for constrained optimization. Appl Soft Comput 38:87–105

Coello CAC, Pulido GT (2001) A micro-genetic algorithm for multiobjective optimization. In: EMO, volume 1993 of lecture notes in computer science, pp 126–140. Springer

Das, K. N. (2014). Hybrid genetic algorithm: an optimization tool. In global trends in intelligent computing Research and Development (pp. 268-305). IGI global.

Das AK, Pratihar DK (2018) A direction-based exponential mutation operator for real-coded genetic algorithm. IEEE International Conference on Emerging Applications of Information Technology.

Dash SR, Dehuri S, Rayaguru S (2013) Discovering interesting rules from biological data using parallel genetic algorithm, 3rd IEEE International Advance Computing Conference (IACC), Ghaziabad,, pp. 631–636.

Datta D, Amaral ARS, Figueira JR (2011) Single row facility layout problem using a permutation-based genetic algorithm. European J Oper Res 213(2):388–394

de Ocampo ALP, Dadios EP (2017) "Energy cost optimization in irrigation system of smart farm by using genetic algorithm," 2017IEEE 9th international conference on humanoid. Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM), Manila, pp 1–7

Deb K, Agrawal RB (1995) Simulated binary crossover for continuous search space. Complex Systems 9:115–148

Deb K, Deb D (2014) Analysing mutation schemes for real-parameter genetic algorithms. International Journal of Artificial Intelligence and Soft Computing 4(1):1–28

Deb K, Pratap A, Agarwal S, Meyarivan T (2002) A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evol Comput 6(2):182–197

Deep K, Das KN (2008) Quadratic approximation based hybrid genetic algorithm for function optimization. Appl Math Comput 203(1):86–98

Deep K, Thakur M (2007) A new mutation operator for real coded genetic algorithms. Appl Math Comput 193:211–230

Deep K, Thakur M (2007) A new crossover operator for real coded genetic algorithms. Appl Math Comput 188:895–911

Dhal KP, Ray S, Das A, Das S (2018) A survey on nature-inspired optimization algorithms and their application in image enhancement domain. Archives of Computational Methods in Engineering 5:1607–1638

Dhiman G, Kumar V (2017) Spotted hyena optimizer: A novel bio-inspired based metaheuristic technique for engineering applications. Adv Eng Softw 114:48–70

Dhiman G, Kumar V (2018) Emperor penguin optimizer: A bio-inspired algorithm for engineering problems. Knowl-Based Syst 159:20–50

Dhiman G, Kumar V (2019) Seagull optimization algorithm: theory and its applications for large-scale industrial engineering problems. Knowl-Based Syst 165:169–196

Di Fatta G, Hoffmann F, Lo Re G, Urso A (2003) A genetic algorithm for the design of a fuzzy controller for active queue management. IEEE Trans Syst Man Cybern Part C Appl Rev 33(3):313–324

Diabat A, Deskoores R (2016) A hybrid genetic algorithm based heuristic for an integrated supply chain problem. J Manuf Syst 38:172–180

Diaz-Manríquez A, Ríos-Alvarado AB, Barrón-Zambrano JH, Guerrero-Melendez TY, Elizondo-Leal JC (2018) An automatic document classifier system based on genetic algorithm and taxonomy. IEEE Access 6:21552–21559. https://doi.org/10.1109/ACCESS.2018.2815992

Article   Google Scholar  

Dorigo M, Birattari M, Stutzle T (2006) Ant colony optimization - artificial ants as a computational intelligence technique. IEEE Comput Intell Mag 1(2006):28–39

Ebrahimzadeh R, Jampour M (2013) Chaotic genetic algorithm based on Lorenz chaotic system for optimization problems. I.J. Intelligent Systems and Applications Intelligent Systems and Applications 05(05):19–24

EkbataniFard GH, Monsefi R, Akbarzadeh-T M-R, Yaghmaee M et al. (2010) A multi-objective genetic algorithm based approach for energy efficient qos-routing in two-tiered wireless sensor net-works. In: wireless pervasive computing (ISWPC), 2010 5th IEEE international symposium on. IEEE, pp 80–85

El-Mihoub T, Hopgood A, Nolle L, Battersby A (2004) Performance of hybrid genetic algorithms incorporating local search. In: Horton G (ed) 18th European simulation multi-conference (ESM2004). Germany, Magdeburg, pp 154–160

El-Mihoub TA, Hopgood AA, Lars N, Battersby A (2006) Hybrid genetic algorithms: A review. Eng Lett 13:2

Emmerich MTM, Deutz AH (2018) A tutorial on multiobjective optimization: fundamentals and evolutionary methods. Nat Comput 17(3):585–609

Eshelman LJ, Caruana RA, Schaffer JD (1997) Biases in the crossover landscape.

Espinoza FB, Minsker B, Goldberg D (2003) Performance evaluation and population size reduction for self adaptive hybrid genetic algorithm (SAHGA), in the Genetic and Evolutionary Computation Conference, vol. 2723, Lecture Notes in Computer Science San Francisco, USA: Springer, pp. 922–933.

Farahani RZ, Elahipanah M (2008) A genetic algorithm to optimize the total cost and service level for just-in-time distribution in a supply chain. Int J Prod Econ 111(2):229–243

Fonseca CM, Fleming PJ (1993) Genetic algorithms for multiobjective optimization: formulation, discussion and generalization. In: ICGA, pp 416–423. Morgan Kaufmann

Fox B, McMahon M (1991) Genetic operators for sequencing problems, in Foundations of Genetic Algorithms, G. Rawlins, Ed. Morgan Kaufmann Publishers, San Mateo,CA, Ed. 1991, pp. 284–300.

Freisleben B, Merz P (1996) New genetic local search operators for the traveling salesman problem," in the Fourth Conference on Parallel Problem Solving from Nature vol. 1141, Lectures Notes in Computer Science, H.-M. Voigt, W. Ebeling, I. Rechenberg, and H.-P. Schwefel, Eds. Berlin, Germany: Springer-Verlag, pp. 890–899.

Friend DH, EI Nainay, M, Shi Y, MacKenzie AB (2008) Architecture and performance of an island genetic algorithm-based cognitive network. In: Consumer communications and networking conference,2008. CCNC 2008. 5th IEEE. IEEE, pp 993–997

Fuertes G, Vargas M, Alfaro M, Soto-Garrido R, Sabattin J, Peralta M-A (2019) Chaotic genetic algorithm and the effects of entropy in performance optimization.

Ghaheri A, Shoar S, Naderan M, Hoseini SS (2015) The applications of genetic algorithms in medicine. CJ 30:406–416

Ghosh S, Bhattachrya S (2020) A data-driven understanding of COVID-19 dynamics using sequential genetic algorithm based probabilistic cellular automata. Applied Soft Computing. 96

Ghoshal AK, Das N, Bhattacharjee S, Chakraborty G (2019) A fast parallel genetic algorithm based approach for community detection in large networks. International Conference on Communication Systems & Networks (COMSNETS), Bengaluru, India, pp. 95–101.

Gogna A, Tayal A (2012) Comparative analysis of evolutionary algorithms for image enhancement. Int J Met 2(1)

Goldberg D (1989) Genetic algorithm in search. Optimization and Machine Learning, Addison -Wesley, Reading, MA 1989

Goldberg D, Lingle R (1985) Alleles, loci and the traveling salesman problem. In: Proceedings of the 1st international conference on genetic algorithms and their applications, vol. 1985. Los Angeles, USA, pp 154–159

Guido R, Conforti D (2017) A hybrid genetic approach for solving an integrated multi-objective operating room planning and scheduling problem. Comput Oper Res 87:270–282

Ha QM, Deville Y, Pham QD, Ha MH (2020) A hybrid genetic algorithm for the traveling salesman problem with drone. J Heuristics 26:219–247

HajiRassouliha A, Gamage TPB, Parker MD, Nash MP, Taberner AJ, Nielsen, PM (2013) FPGA implementation of 2D cross-correlation for real-time 3D tracking of deformable surfaces. In Proceedings of the2013 28th International Conference on Image and Vision Computing New Zealand (IVCNZ 2013), Wellington, New Zealand, 27–29 November 2013; IEEE: Piscataway, NJ, USA; pp. 352–357

Harada T, Alba E (2020) Parallel genetic algorithms: a useful survey. ACM Computing Survey 53(4):1–39

Harik GR, Lobo FG (1999) A parameter-less genetic algorithm, in Proceedings of the Genetic and Evolutionary Computation Conference, pp. 258–265.

Hassanat A, Almohammadi K, Alkafaween E, Abunawas E, Hammouri A, Prasath VBS (December 2019) Choosing mutation and crossover ratios for genetic algorithms—A review with a new dynamic approach. Information 10:390. https://doi.org/10.3390/info10120390

He J, Ji S, Yan M, Pan Y, Li Y (2012) Load-balanced CDS construction in wireless sensor networks via genetic algorithm. Int J Sens Netw 11(3):166–178

Hedar A, Fukushima M (2003) Simplex coding genetic algorithm for the global optimization of nonlinear functions, in Multi-Objective Programming and Goal Programming, Advances in Soft Computing, T. Tanino, T. Tanaka, and M. Inuiguchi, Eds.: Springer-Verlag, pp. 135–140.

Helal MHS, Fan C, Liu D, Yuan S (2017) Peer-to-peer based parallel genetic algorithm. International Conference on Information, Communication and Engineering (ICICE), Xiamen, pp 535–538

Hiassat A, Diabat A, Rahwan I (2017) A genetic algorithm approach for location-inventory-routing problem with perishable products. J Manuf Syst 42:93–103

Holland JH (1975) Adaptation in natural and artificial systems. The U. of Michigan Press

Hong W-C, Dong Y, Chen L-Y, Wei S-Y (2011) SVR with hybrid chaotic genetic algorithms for tourism demand forecasting. Appl Soft Comput 11(2):1881–1890

Hong T-P, Lee Y-C, Min-Thai W (2014) An effective parallel approach for genetic-fuzzy data mining. Exp Syst Applic 41(2):655–662

Horn J, Nafpliotis N, Goldberg DE. (1994) A niched Pareto genetic algorithm for multiobjective optimization. Proceedings of the First IEEE Conference on Evolutionary Computation, IEEE World Congress on Computational Intelligence, vol. 1, Piscataway, NJ: IEEE Service Center, p. 67–72.

Hu C, Wang X, Mandal MK, Meng M, Li D (2003) Efficient face and gesture recognition techniques for robot control. Department of Electrical and Computer Engineering University of Alberta, Edmonton, AB, T6G 2V4, Canada. CCECE2003 - CCGEI 2003, Montreal, May/mai 2003 IEEE, pp 1757-1762.

Peng Huo, Simon C. K. Shiu, Haibo Wang, Ben Niu (2009) Application and Comparison of Particle Swarm Optimization and Genetic Algorithm in Strategy Defense Game. Fifth International Conference on Natural Computation, pp 387–392.

Hussain A, Muhammad YS, Nauman Sajid M, Hussain I, Mohamd Shoukry A, Gani S (2017) Genetic algorithm for traveling salesman problem with modified cycle crossover operator. Computational intelligence and neuroscience 2017:1–7

Ishibuchi H, Murata T (1998) A multi-objective genetic local search algorithm and its application to flowshop scheduling. IEEE Trans Syst Man Cybern Part C Appl Rev 28(3):392–403

Jafari A, Khalili T, Babaei E, Bidram A (2020) Hybrid optimization technique using exchange market and GA. IEEE Access 8:2417–2427

Jaszkiewicz A (February 2002) Genetic local search for multi-objective combinatorial optimization. Eur J Oper Res 137(1):50–71

Javidi M, Hosseinpourfard R (2015) Chaos genetic algorithm instead genetic algorithm. Int J Inf Tech 12(2):163–168

Jebari K (2013) Selection methods for genetic algorithms. Abdelmalek Essaâdi University. International Journal of Emerging Sciences 3(4):333–344

Jiang S, Chin K-S, Wang L, Qu G, Tsui KL (2017) Modified genetic algorithm-based feature selection combined with pre-trained deep neural network for demand forecasting in outpatient department. Expert Syst Appl 82:216–230

Jiang M, Fan X, Pei Z, Zhang Z (2018) Research on text feature clustering based on improved parallel genetic algorithm. Tenth International Conference on Advanced Computational Intelligence (ICACI), Xiamen, pp. 235–238

Kaluri R, Reddy P (2016) Sign gesture recognition using modified region growing algorithm and adaptive genetic fuzzy classifier. International Journal of Intelligent Engineering and Systems 9(4):225–233

Kandavanam G, Botvich D, Balasubramaniam S, Jennings B (2010) A hybrid genetic algorithm/variable neighborhood search approach to maximizing residual bandwidth of links for route planning. Artificial evolution. Springer, In, pp 49–60

Kannan S (2020) Intelligent object recognition in underwater images using evolutionary-based Gaussian mixture model and shape matching. SIViP 14:877–885

Karabudak D, Hung C-C, Bing B (2004) A call admission control scheme using genetic algorithms. In: Proceedings of the 2004ACM symposium on applied computing. ACM, pp 1151–1158

Katz P, Aron M, Alfalou A (2001) A face-tracking system to detect falls in the elderly; SPIE newsroom. SPIE, Bellingham, WA, USA, p 201

Kaur M, Kumar V (2018) Beta chaotic map based image encryption using genetic algorithm. Int J Bifurcation Chaos 28(11):1850132

Kaur M, Kumar V (2018) Parallel non-dominated sorting genetic algorithm-II-based image encryption technique. The Imaging Science Journal. 66(8):453–462

Kaur M, Kumar V (2018) Fourier–Mellin moment-based intertwining map for image encryption. Modern Physics Letters B 32(9):1850115

Kaur G, Bhardwaj N, Singh PK (2018) An analytic review on image enhancement techniques based on soft computing approach. Sensors and Image Processing, Advances in Intelligent Systems and Computing 651:255–266

Kavitha AR, Chellamuthu C (2016) Brain tumour segmentation from MRI image using genetic algorithm with fuzzy initialisation and seeded modified region growing (GFSMRG) method. The Imaging Science Journal 64(5):285–297

Kennedy J, Eberhart RC (1995) Particle swarm optimization. In: Proceedings of IEEE international conference on neural networks (1995), pp 1942–1948

Khan, A., ur Rehman, Z., Jaffar, M.A., Ullah, J., Din, A., Ali, A., Ullah, N. (2019) Color image segmentation using genetic algorithm with aggregation-based clustering validity index (CVI). SIViP 13(5), 833–841

Kia R, Khaksar-Haghani F, Javadian N, Tavakkoli-Moghaddam R (2014) Solving a multi-floor layout design model of a dynamic cellular manufacturing system by an efficient genetic algorithm. J Manuf Syst 33(1):218–232

Kim EY, Jung K (2006) Genetic algorithms for video segmentation. Pattern Recogn 38(1):59–73

Kim EY, Park SH (2006) Automatic video segmentation using genetic algorithms. Pattern Recogn Lett 27(11):1252–1265

Kita H, Ono I, Kobayashi S (1999). The multi-parent unimodal normal distribution crossover for real-coded genetic algorithms. Proceedings of the 1999 Congress on Evolutionary Computation, vol. 2, IEEE (1999), pp. 1588–1595

Kobayashi H, Munetomo M, Akama K, Sato Y (2004) Designing a distributed algorithm for bandwidth allocation with a genetic algorithm. Syst Comput Jpn 35(3):37–45

Konak A, Smith AE (1999) A hybrid genetic algorithm approach for backbone design of communication networks, in the 1999 Congress on Evolutionary Computation. Washington D.C, USA: IEEE, pp. 1817-1823.

Kortil Y, Jridi M, Falou AA, Atri M (2020) Face recognition systems: A survey. Sensors. 20:1–34

Krishnan N, Muthukumar S, Ravi S, Shashikala D, Pasupathi P (2013) Image restoration by using evolutionary technique to Denoise Gaussian and impulse noise. In: Prasath R., Kathirvalavakumar T. (eds) mining intelligence and knowledge exploration. Lecture notes in computer science, vol 8284. Springer, Cham.

Kumar A (2013) Encoding schemes in genetic algorithm. Int J Adv Res IT Eng 2(3):1–7

Kumar V, Kumar D (2017) An astrophysics-inspired grey wolf algorithm for numerical optimization and its application to engineering design problems. Adv Eng Softw 112:231–254

Kumar V, Chhabra JK, Kumar D (2014) Parameter adaptive harmony search algorithm for unimodal and multimodal optimization problems. J Comput Sci 5(2):144–155

Kumar C, Singh AK, Kumar P (2017) A recent survey on image watermarking techniques and its application in e-governance. MultiMed Tools Appl.

Kurdi M (2016) An effective new island model genetic algorithm for job shop scheduling problem. Comput Oper Res 67(2016):132–142

Larranaga P, Kuijpers CMH, Murga RH, Yurramendi Y (July 1996) Learning Bayesian network structures by searching for the best ordering with genetic algorithms. in IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans 26(4):487–493

Larranaga P, Kuijpers C, Murga R, Inza I, Dizdarevic S (1999) Genetic algorithms for the travelling salesman problem: a review of representations and operators. Artificial Intelligence Review 13:129–170

Chang-Yong Lee (2003) Entropy-Boltzmann selection in the genetic algorithms. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), vol. 33, no. 1, pp. 138–149, Feb. 2003.

Lee CKH (2018) A review of applications of genetic algorithms in operations management. Eng Appl Artif Intell 76:1–12

Lee Y, Hara T, Fujita H, Itoh S, Ishigaki T (July 2001) Automated detection of pulmonary nodules in helical CT images based on an improved template-matching technique. in IEEE Transactions on Medical Imaging 20(7):595–604

Joon-Yong Lee, Min-Soeng Kim, Cheol-Taek Kim and Ju-Jang Lee (2007) Study on encoding schemes in compact genetic algorithm for the continuous numerical problems,SICE Annual Conference 2007, Takamatsu, pp. 2694–2699.

Leng LT (1999) Guided genetic algorithm. University of Essex, Doctoral Dissertation

Li B, Li J, Tang K, Yao X (2015) Many-objective evolutionary algorithms: A survey. ACM Computing surveys

Lie Tang L (2000) Tian and Brian L steward, "color image segmentation with genetic algorithm for in-field weed sensing". Transactions of the ASAE 43(4):1019–1027

Lima S.J.A., de Araújo S.A. (2018) A new binary encoding scheme in genetic algorithm for solving the capacitated vehicle routing problem. In: Korošec P., Melab N., Talbi EG. (eds) Bioinspired Optimization Methods and Their Applications. BIOMA 2018. Lecture notes in computer science, vol 10835. Springer, Cham

Liu D (2019) Mathematical modeling analysis of genetic algorithms under schema theorem. Journal of Computational Methods in Sciences and Engineering 19:S131–S137

Liu Z, Meng Q, Wang S (2013) Speed-based toll design for cordon-based congestion pricing scheme. Transport Res Part C: Emerg Technol 31(2013):83–98

Lorenzo B, Glisic S (2013) Optimal routing and traffic scheduling for multihop cellular networks using genetic algorithm. IEEE Trans Mob Comput 12(11):2274–2288

Lucasius CB, Kateman G (1989) Applications of genetic algorithms in chemometrics. In: Proceedings of the 3rd international conference on genetic algorithms. Morgan Kaufmann, Los Altos, CA, USA, pp 170–176

Luo B, Jinhua Zheng, Jiongliang Xie, Jun Wu. Dynamic crowding distance – a new diversity maintenance strategy for MOEAs. ICNC ‘08, Fourth Int. Conf. on Natural Comp., vol. 1 (2008), pp. 580–585

Maghawry A, Kholief M, Omar Y, Hodhod R (2020) An approach for evolving transformation sequences using hybrid genetic algorithms. Int J Intell Syst 13(1):223–233

Manzoni L, Mariot L, Tuba E (2020) Balanced crossover operators in genetic algorithms. Swarm and Evolutionary Computation 54:100646

Mazinani M, Abedzadeh M, Mohebali N (2013) Dynamic facility layout problem based on flexible bay structure and solving by genetic algorithm. Int J Adv Manuf Technol 65(5–8):929–943

Mehboob U, Qadir J, Ali S, Vasilakos A (2016) Genetic algorithms in wireless networking: techniques, applications, and issues. Soft Comput 20:2467–2501

Michalewicz Z (1992) Genetic algorithms + data structures = evolution programs. Springer-Verlag, New York

Michalewicz Z, Schoenauer M (1996) Evolutionary algorithms for constrained parameter optimization problems. Evol Comput 4(1):1–32

Mishra R, Das KN (2017). A novel hybrid genetic algorithm for unconstrained and constrained function optimization. In bio-inspired computing for information retrieval applications (pp. 230-268). IGI global

Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6(7):e1000097

Mooi S, Lim S, Sultan M, Bakar A, Sulaiman M, Mustapha A, Leong KY (2017) Crossover and mutation operators of genetic algorithms. International Journal of Machine Learning and Computing 7:9–12

Mudaliar DN, Modi NK (2013) Unraveling travelling salesman problem by genetic algorithm using m-crossover operator. International Conference on Signal Processing, Image Processing & Pattern Recognition, Coimbatore, pp 127–130

T. Murata and M. Gen (2000) Cellular genetic algorithm for multi-objective optimization, in Proceedings of the Fourth Asian Fuzzy System Symposium, pp. 538–542

Neto JC, Meyer GE, Jones DD (2006) Individual leaf extractions from young canopy images using gustafsonkessel clustering and a genetic algorithm. Comput Electron Agric 51(1):66–85

NKFC, Viswanatha SDK (2009) Routing algorithm using mobile agents and genetic algorithm. Int J Comput Electr Eng, vol 1, no 3

Ono I, Kobayashi S (1997) A real-coded genetic algorithm for functional optimization using unimodal normal distribution crossover. In: Back T (ed) Proceedings of the 7th international conference on genetic algorithms, ICGA-7. Morgan Kaufmann, East Lansing, MI, USA, pp 246–253

Pachepsky Y, Acock B (1998) Stochastic imaging of soil parameters to assess variability and uncertainty of crop yield estimates. Geoderma 85(2):213–229

Paiva JPD, Toledo CFM, Pedrini H (2016) An approach based on hybrid genetic algorithm applied to image denoising problem. Appl Soft Comput 46:778–791

Palencia AER, Delgadillo GEM (2012) A computer application for a bus body assembly line using genetic algorithms. Int J Prod Econ 140(1):431–438

Palomo-Romero JM, Salas-Morera L, García-Hernández L (2017) An island model genetic algorithm for unequal area facility layout problems. Expert Syst Appl 68:151–162

Pandian S, Modrák V (December 2009) "possibilities, obstacles and challenges of genetic algorithm in manufacturing cell formation," advanced logistic systems, University of Miskolc. Department of Material Handling and Logistics 3(1):63–70

Park Y-B, Yoo J-S, Park H-S (2016) A genetic algorithm for the vendor-managed inventory routing problem with lost sales. Expert Syst Appl 53:149–159

Patel R, Raghuwanshi MM, Malik LG (2012) Decomposition based multi-objective genetic algorithm (DMOGA) with opposition based learning

Pattanaik JK, Basu M, Dash DP (2018) Improved real coded genetic algorithm for dynamic economic dispatch. Journal of electrical systems and information technology. Vol. 5(3):349–362

Payne AW, Glen RC (1993) Molecular recognition using a binary genetic system. J Mol Graph 11(2):74–91

Peerlinck A, Sheppard J, Pastorino J, Maxwell B (2019) Optimal Design of Experiments for precision agriculture using a genetic algorithm. IEEE Congress on Evolutionary Computation.

Pelikan M, Goldberg DE, Cantu-Paz E (2000) Bayesian optimization algorithm, population sizing, and time to convergence, Illinois Genetic Algorithms Laboratory, University of Illinois, Tech. Rep

Pilat ML, White T (2002) Using genetic algorithms to optimize ACS-TSP, in the Third International Workshop on Ant Algorithms, vol. Lecture Notes In Computer Science 2463. Berlin, Germany: Springer-Verlag, pp. 282–287.

Pinagapany S, Kulkarni A (2008) Solving channel allocation problem in cellular radio networks using genetic algorithm. In: Communication Systems software and middleware and workshops, 2008.COMSWARE 2008. 3rd International Conference on. IEEE, pp239–244

Pinel F, Dorronsoro B, Bouvry P (2013) Solving very large instances of the scheduling of independent tasks problem on the GPU. J Parallel Distrib. Comput 73(1):101–110

Pinto G, Ainbinder I, Rabinowitz G (2009) A genetic algorithm-based approach for solving the resource-sharing and scheduling problem. Comput Ind Eng 57(3):1131–1143

Piszcz A, Soule T (2006) Genetic programming: optimal population sizes for varying complexity problems, in Proceedings of the Genetic and Evolutionary Computation Conference, pp. 953–954.

Porta J, Parapar R, Doallo F, Rivera F, Santé I, Crecente R (2013) High performance genetic algorithm for land use planning. Comput Environ Urb Syst 37(2013):45–58

Rafsanjani MK, Riyahi M (2020) A new hybrid genetic algorithm for job shop scheduling problem. International Journal of Advanced Intelligence Paradigms 16(2):157–171

Rathi R, Acharjya DP (2018) A framework for prediction using rough set and real coded genetic algorithm. Arab J Sci Eng 43(8):4215–4227

Rathi R, Acharjya DP (2018) A rule based classification for vegetable production using rough set and genetic algorithm. International Journal of Fuzzy System Applications (IJFSA) 7(1):74–100

Rathi R, Acharjya DP (2020) A comparative study of genetic algorithm and neural network computing techniques over feature selection, In advances in distributed computing and machine learning (pp. 491–500). Springer, Singapore

Ray SS, Bandyopadhyay S, Pal SK (2004) New operators of genetic algorithms for traveling salesman problem," Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004., Cambridge pp 497-500

Richter JN, Peak D (2002) Fuzzy evolutionary cellular automata, in international conference on artificial neural networks in engineering, vol 12. USA, Saint Louis pp. 185-191

Riedl A (2002) A hybrid genetic algorithm for routing optimization in ip networks utilizing bandwidth and delay metrics. In: IP operations and management, 2002 IEEE Workshop on. IEEE, pp 166–170

Ripon KSN, Siddique N, Torresen J (2011) Improved precedence preservation crossover for multi-objective job shop scheduling problem. Evolving Systems 2:119–129

Roberge V, Tarbouchi M, Okou F (2014) Strategies to accelerate harmonic minimization in multilevel inverters using a parallel genetic algorithm on graphical processing unit. IEEE Trans Power Electron 29(10):5087–5090

Ronald S (1997) Robust encoding in genetic algorithms: a survey of encoding issues. IEEE international conference on evolutionary computation, pp. 43-48

Roy A, Banerjee N, Das SK (2002) An efficient multi-objective qos-routing algorithm for wireless multicasting. In:Vehicular technology conference, 2002. VTC Spring 2002. IEEE 55th, vol 3., pp 1160–1164

Sadrzadeh A (2012) A genetic algorithm with the heuristic procedure to solve the multi-line layout problem. Comput Ind Eng 62(4):1055–1064

Sahingoz OK (2014) Generation of Bezier curve-based flyable trajectories for multi-UAV systems with parallel genetic algorithm. J Intell Robot Syst 74(1):499–511

Saini N (2017) Review of selection methods in genetic algorithms. International Journal of Engineering and Computer Science 6(12):22261–22263

Sari M, Can T (2018) Prediction of pathological subjects using genetic algorithms. Computational and Mathematical Methods in Medicine 2018:1–9

Scully T, Brown KN (2009) Wireless LAN load balancing with genetic algorithms. Knowl Based Syst 22(7):529–534

Sermpinis G, Stasinakis C, Theofilatos K, Karathanasopoulos A (2015) Modeling, forecasting and trading the EUR exchange rates with hybrid rolling genetic algorithms–support vector regression forecast combinations. European J. Oper. Res. 247(3):831–846

Shabankareh SG, Shabankareh SG (2019) Improvement of edge-tracking methods using genetic algorithm and neural network, 2019 5th Iranian conference on signal processing and intelligent systems (ICSPIS). Shahrood, Iran, pp 1–7. https://doi.org/10.1109/ICSPIS48872.2019.9066026

Book   Google Scholar  

Sharma S, Gupta K (2011) Solving the traveling salesman problem through genetic algorithm with new variation order crossover. International Conference on Emerging Trends in Networks and Computer Communications (ETNCC), Udaipur, pp. 274–276

Sharma N, Kaushik I, Rathi, R, Kumar S (2020) Evaluation of accidental death records using hybrid genetic algorithm. Available at SSRN: https://ssrn.com/abstract=3563084 or https://doi.org/10.2139/ssrn.3563084

Shayeghi A, Gotz D, Davis JBA, Schafer R, Johnston RL (2015) Pool-BCGA: A parallelised generation-free genetic algorithm for the ab initio global optimisation of nano alloy clusters. Phys Chem Chem Phys 17(3):2104–2112

Guoyong Shi, H. Iima and N. Sannomiya (1996) A new encoding scheme for solving job shop problems by genetic algorithm, Proceedings of 35th IEEE Conference on Decision and Control, Kobe, Japan, 1996, pp. 4395–4400 vol.4.

Shi J, Liu Z, Tang L, Xiong J (2017) Multi-objective optimization for a closed-loop network design problem using an improved genetic algorithm. Appl Math Model 45:14–30

Shukla AK, Singh P, Vardhan M (2019) A new hybrid feature subset selection framework based on binary genetic algorithm and information theory. International Journal of Computational Intelligence and Applications 18(3):1950020(1–10)

Singh A, Deep K (2015) Real coded genetic algorithm operators embedded in gravitational search algorithm for continuous optimization. Int J Intell Syst Appl 7(12):1

Sivanandam SN, Deepa SN (2008) Introduction to genetic algorithm, 1st edn. Springer-Verlag, Berlin Heidelberg

Soleimani H, Kannan G (2015) A hybrid particle swarm optimization and genetic algorithm for closed-loop supply chain network design in large-scale networks. Appl Math Model 39(14):3990–4012

Soleimani H, Govindan K, Saghafi H, Jafari H (2017) Fuzzy multi-objective sustainable and green closed-loop supply chain network design. Comput Ind Eng 109:191–203

Soon GK, Guan TT, On CK, Alfred R, Anthony P (2013) "A comparison on the performance of crossover techniques in video game," 2013 IEEE international conference on control system. Computing and Engineering, Mindeb, pp 493–498

Srinivas N, Deb K (1995) Multi-objective function optimization using non-dominated sorting genetic algorithms. Evol Comput 2(3):221–248

Subbaraj P, Rengaraj R, Salivahanan S (2011) Enhancement of self-adaptive real-coded genetic algorithm using Taguchi method for economic dispatch problem. Appl Soft Comput 11(1):83–92

Tahir M, Tubaishat A, Al-Obeidat F, et al. (2020) A novel binary chaotic genetic algorithm for feature selection and its utility in affective computing and healthcare. Neural Comput & Appl

Tam V, Cheng K-Y, Lui K-S (2006) Using micro-genetic algorithms to improve localization in wireless sensor networks. J Commun 1(4):1–10

Tan KC, Li Y, Murray-Smith DJ, Sharman KC (1995) System identification and linearisation using genetic algorithms with simulated annealing, in First IEE/IEEE Int. Conf. on GA in Eng. Syst.: Innovations and Appl. Sheffield, UK, pp. 164–169.

Tang PH, Tseng MH (2013) Adaptive directed mutation for real-coded genetic algorithms. Appl Soft Comput 13(1):600–614

Tiong SK, Yap DFW, Koh SP (2012) A comparative analysis of various chaotic genetic algorithms for multimodal function optimization. Trends in Applied Sciences Research 7:785–791

Toutouh J, Alba E (2017) Parallel multi-objective metaheuristics for smart communications in vehicular networks. Soft Comput 21(8):1949–1961

Umbarkar A, Sheth P (2015) Crossover operators in genetic algorithms: a review. Journal on Soft Computing 6(1)

Verma D, Vishwakarma VP, Dalal S (2020) A hybrid self-constrained genetic algorithm (HSGA) for digital image Denoising based on PSNR improvement. Advances in Bioinformatics, Multimedia, and Electronics Circuits and Signals, In, pp 135–153

Vitayasak S, Pongcharoen P, Hicks C (2016) A tool for solving stochastic dynamic facility layout problems with stochastic demand using either a genetic algorithm or modified backtracking search algorithm. Int J Prod Econ

Junru Wang and Lan Huang (2014) Evolving gomoku Solver by Genetic Algorithm. IEEE Workshop on Advanced Research and Technology in Industry Applications (WARTIA) pp 1064–1067.

Wang L, Kan MS, Shahriar Md R, Tan ACC (2014) Different approaches of applying single-objective binary genetic algorithm on the wind farm design. In World Congress on Engineering Asset Management.

Wang N, Li Q, Abd El-Latif AA, Zhang T, Niu X (2014) Toward accurate localization and high recognition performance for noisy iris images. Multimed Tools Appl 71(3):1411–1430

Wang JQ, Ersoy OK, He MY et al (2016) Multi-offspring genetic algorithm and its application to the traveling salesman problem. Appl Soft Comput 43:415–423

Wang FL, Fu XM, Zhu HX et al (2016) Multi-child genetic algorithm based on two-point crossover. J Northeast Agric Univ 47(3):72–79

Wang JQ, Cheng ZW, Ersoy OK et al (2018) Improvement analysis and application of real-coded genetic algorithm for solving constrained optimization problems. Math Probl Eng 2018:1–16

Wang J, Zhang M, Ersoy OK, Sun K, Bi Y (2019) An improved real-coded genetic algorithm using the Heuristical Normal distribution and direction-based crossover. Computational Intelligence and Neuroscience 2019:1–17

Wen Z, Yang R, Garraghan P, Lin T, Xu J, Rovatsos M (2017) Fog orchestration for internet of things services. IEEE Internet Comput 21(2) (Mar. 2017):16–24

Wright AH (1991) Genetic algorithms for real parameter optimization. In Foundations of genetic algorithms I,G. J. E. Rawlins, Ed., Morgan Kaufmann, San Mateo, CA,USA

Wu X, Chu C-H, Wang Y, Yan W (2007) A genetic algorithm for cellular manufacturing design and layout. European J Oper Res 181(1):156–167

Yang S, Cheng H, Wang F (2010) Genetic algorithms with immigrants and memory schemes for dynamic shortest path routing problems in mobile ad hoc networks. IEEE Trans Syst Man Cybern Part C Appl Rev 40(1):52–63

Yang C, Li H, Rezgui Y, Petri I, Yuce B, Chen B, Jayan B (2014) High throughput computing based distributed genetic algorithm for building energy consumption optimization. Energy Build 76(2014):92–101

Yu F, Xu X (2014) A short-term load forecasting model of natural gas based on optimized genetic algorithm and improve BR neural network. Appl Energy 134:102–113

Yuce B, Fruggiero F, Packianather MS, Pham DT, Mastrocinque E, Lambiase A, Fera M (2017) Hybrid genetic bees algorithm applied to single machine scheduling with earliness and tardiness penalties. Comput Ind Eng 113:842–858

Yun S, Lee J, Chung W, Kim E, Kim S (2009) A soft computing approach to localization in wireless sensor networks. Expert Syst Appl 36(4):7552–7561

Zhai R (2020) Solving the optimization of physical distribution routing problem with hybrid genetic algorithm. J Phys Conf Ser 1550:1–6

Zhang Q, Wang J, Jin C, Zeng Q (2008) Localization algorithm for wireless sensor network based on genetic simulated annealing algorithm. In: 4th IEEE International Conference on Wireless communications, networking and mobile computing. Pp 1–5

Zhang R, Ong SK, Nee AYC (2015) A simulation-based genetic algorithm approach for remanufacturing process planning and scheduling. Appl Soft Comput 37:521–532

Zhang X-Y, Zhang J, Gong Y-J, Zhan Z-H, Chen W-N, Li Y (2016) Kuhn-Munkres parallel genetic algorithm for the set cover problem and its application to large-scale wireless sensor networks. IEEETrans Evol Comput 20(5):695–710

Zhenhua Y, Guangwen Y, Shanwei L, Qishan Z (2010) A modified immune genetic algorithm for channel assignment problems in cellular radio networks. In: Intelligent system design and engineering application (ISDEA), 2010 International Conference on, vol 2. , pp 823–826

Download references

Author information

Authors and affiliations.

Computer Science and Engineering Department, National Institute of Technology, Hamirpur, India

Sourabh Katoch, Sumit Singh Chauhan & Vijay Kumar

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Vijay Kumar .

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Katoch, S., Chauhan, S.S. & Kumar, V. A review on genetic algorithm: past, present, and future. Multimed Tools Appl 80 , 8091–8126 (2021). https://doi.org/10.1007/s11042-020-10139-6

Download citation

Received : 27 July 2020

Revised : 12 October 2020

Accepted : 23 October 2020

Published : 31 October 2020

Issue Date : February 2021

DOI : https://doi.org/10.1007/s11042-020-10139-6

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Optimization
• Metaheuristic
• Genetic algorithm
• Find a journal
• Publish with us
• Track your research

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

•  We're Hiring!
•  Help Center
• Most Cited Papers
• Most Downloaded Papers
• Newest Papers
• Save to Library
• Last »
• Genomics Follow Following
• Bioinformatics Follow Following
• Molecular Biology Follow Following
• Evolutionary Biology Follow Following
• Population Genetics Follow Following
• Molecular Genetics Follow Following
• Biology Follow Following
• Plant breeding and genetics Follow Following
• Evolutionary genetics Follow Following
• Plant Breeding Follow Following

Enter the email address you signed up with and we'll email you a reset link.

• Academia.edu Publishing
•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

• Alzheimer's disease & dementia
• Arthritis & Rheumatism
• Attention deficit disorders
• Autism spectrum disorders
• Biomedical technology
• Diseases, Conditions, Syndromes
• Endocrinology & Metabolism
• Gastroenterology
• Gerontology & Geriatrics
• Health informatics
• Inflammatory disorders
• Medical economics
• Medical research
• Medications
• Neuroscience
• Obstetrics & gynaecology
• Oncology & Cancer
• Ophthalmology
• Overweight & Obesity
• Parkinson's & Movement disorders
• Psychology & Psychiatry
• Radiology & Imaging
• Sleep disorders
• Sports medicine & Kinesiology
• Vaccination
• Breast cancer
• Cardiovascular disease
• Chronic obstructive pulmonary disease
• Colon cancer
• Coronary artery disease
• Heart attack
• Heart disease
• High blood pressure
• Kidney disease
• Lung cancer
• Multiple sclerosis
• Myocardial infarction
• Ovarian cancer
• Post traumatic stress disorder
• Rheumatoid arthritis
• Schizophrenia
• Skin cancer
• Type 2 diabetes
• Full List »

share this!

March 27, 2024

This article has been reviewed according to Science X's editorial process and policies . Editors have highlighted the following attributes while ensuring the content's credibility:

fact-checked

peer-reviewed publication

trusted source

Higher genetic risk of obesity means working out harder for same results, study shows

by Vanderbilt University Medical Center

Persons with a higher genetic risk of obesity need to work out harder than those of moderate or low genetic risk to avoid becoming obese, according to a Vanderbilt University Medical Center (VUMC) paper published in JAMA Network Open.

Study authors used activity, clinical and genetic data from the National Institutes of Health's "All of Us" Research Program to explore the association of genetic risk of higher body mass index and the level of physical activity needed to reduce incident obesity .

"Physical activity guidelines do not account for individual differences ," said senior author Douglas Ruderfer, Ph.D., associate professor of Medicine, Division of Genetic Medicine, and director of the Center for Digital Genomic Medicine at VUMC. "Genetic background contributes to the amount of physical activity needed to mitigate obesity. The higher the genetic risk, the more steps needed per day."

"I think an important component to this result is that individuals can be active enough to account for their genetic background, or their genetic risk for obesity, regardless of how high that risk might be," he added. "And there are many other contributors that play a role including diet and environmental factors."

Included in the study were 3,124 middle-aged participants without obesity who owned a Fitbit device and walked an average of 8,326 steps per day for a median of more than five years. The incidence of obesity over the study period increased from 13% to 43% in the lowest and highest polygenic risk score groups.

Individuals with a polygenic risk score in the 75th percentile would need to walk an average of 2,280 more steps per day (a total of 11,020 steps per day) than those in the 50th percentile to have a comparable risk of obesity, according to the study.

Persons with a baseline BMI of 22, 24, 26 and 28 who were in the 75th percentile of polygenic risk score would need to walk an additional 3,460, 4,430, 5,380 and 6,350 steps per day, respectively, to have a comparable risk of obesity to persons in the 25th percentile.

"I think it is intuitive that individuals who have a higher genetic risk of obesity might need to have more physical activity to reduce that risk, but what is new and important from this study is that we were able to put a number on the amount of activity needed to reduce the risk," said lead author Evan Brittain, MD, associate professor of Medicine in the Division of Cardiovascular Medicine at VUMC and lead investigator in Digital Health for the All of Us Research Program Data and Research Center.

"It is becoming more commonplace to know you have a genetic risk for obesity in the genomic era when genetic results are being returned directly to patients. And you can imagine a future in which that data could be integrated with someone's electronic health record and could form the basis of an individual's physical activity recommendation from their doctor."

Study authors said they now want to see if the findings generalize to more representative and diverse populations in order to determine if providing information for individual activity recommendations results in improved health and a lower likelihood of obesity.

"We would like to test whether knowledge of one's genetic risk for obesity actually has an impact on their behavior," Brittain said. "I think these findings could be empowering for patients because the current physical activity guidelines take a one-size-fits-all approach, and what we learned is that depending on your genetic risk, the guidelines may underestimate the amount of activity needed to reduce your risk of obesity.

"Most importantly, I would like for patients to know that your genetic risk doesn't determine your overall risk of obesity, and you can actually overcome that risk by being more active," he added.

Explore further

Feedback to editors

'Zombie neurons' shed light on how the brain learns

Person is diagnosed with bird flu after being in contact with cows in Texas

14 hours ago

'Pathogen prospecting': Mosquito researchers track malaria's history by examining its epidemiology

Pilot study shows ketogenic diet improves severe mental illness

Chatbot outperforms physicians in clinical reasoning, but also underperforms against residents on many occasions

Reducing late-night alcohol sales curbed all violent crimes by 23% annually in a Baltimore neighborhood: Study

Scientists pioneer immunotherapy technique for autoimmune diseases

Researchers develop more broadly protective coronavirus vaccine

15 hours ago

Scientists discover speed of visual perception ranges widely in humans

16 hours ago

Move more, sleep better: Study finds physical activity lengthens REM latency

17 hours ago

Related Stories

People whose genotype supports physical activity found to have lower risk of developing cardiovascular disease

Nov 6, 2023

People with genetic predisposition to obesity may have lower risk of cardiovascular disease

Apr 6, 2023

Genetic information can improve colorectal cancer screening, shows study

Jan 17, 2024

Obesity is a critical risk factor for type 2 diabetes, regardless of genetics

Apr 15, 2020

Investigating the impact of replacing sedentary time with physical activity on genetic risk of coronary heart disease

Oct 23, 2023

Physical inactivity and restless sleep exacerbate genetic risk of obesity

Oct 20, 2017

Recommended for you

International study uses AI to show how personality influences the expression of our genes

19 hours ago

New gene discovery leads advance against a form of heart failure prevalent in men

Genetic causes of cerebral palsy uncovered through whole-genome sequencing

Mar 29, 2024

Veterans help provide greater insight into Klinefelter and Jacobs syndromes

Research suggests fine-tuning of specific excitatory synapse traits could lead to new brain disease treatments

Women with obesity do not need to gain weight during pregnancy, new study suggests

Mar 28, 2024

Let us know if there is a problem with our content

Use this form if you have come across a typo, inaccuracy or would like to send an edit request for the content on this page. For general inquiries, please use our contact form . For general feedback, use the public comments section below (please adhere to guidelines ).

Please select the most appropriate category to facilitate processing of your request

Thank you for taking time to provide your feedback to the editors.

Your feedback is important to us. However, we do not guarantee individual replies due to the high volume of messages.

E-mail the story

Your email address is used only to let the recipient know who sent the email. Neither your address nor the recipient's address will be used for any other purpose. The information you enter will appear in your e-mail message and is not retained by Medical Xpress in any form.

Newsletter sign up

Get weekly and/or daily updates delivered to your inbox. You can unsubscribe at any time and we'll never share your details to third parties.

More information Privacy policy

Donate and enjoy an ad-free experience

We keep our content available to everyone. Consider supporting Science X's mission by getting a premium account.

E-mail newsletter

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List

Celebrating a Century of Research in Behavioral Genetics

Robert plomin.

Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, UK

A century after the first twin and adoption studies of behavior in the 1920s, this review looks back on the journey and celebrates milestones in behavioral genetic research. After a whistle-stop tour of early quantitative genetic research and the parallel journey of molecular genetics, the travelogue focuses on the last fifty years. Just as quantitative genetic discoveries were beginning to slow down in the 1990s, molecular genetics made it possible to assess DNA variation directly. From a rocky start with candidate gene association research, by 2005 the technological advance of DNA microarrays enabled genome-wide association studies, which have successfully identified some of the DNA variants that contribute to the ubiquitous heritability of behavioral traits. The ability to aggregate the effects of thousands of DNA variants in polygenic scores has created a DNA revolution in the behavioral sciences by making it possible to use DNA to predict individual differences in behavior from early in life.

Introduction

Although the history of heredity and behavior can be traced back to ancient times (Loehlin 2009 ), the first human behavioral genetic research was reported in the 1920s, which applied quantitative genetic twin and adoption designs to assess genetic influence on newly developed measures of intelligence. The 1920s also marked the beginning of single-gene research that led to molecular genetics. The goal of this review is to outline 100 years of progress in quantitative genetic and molecular genetic research on behavior, a whistle-stop tour of a few of the major milestones in the journey. The review focuses on human research even though non-human animal research played a major role in the first 50 years (Maxson 2007 ). It uses intelligence as a focal example because intelligence was the target of much human research, even though a similar story could be told for other areas of behavioral genetics such as psychopathology.

The Two Worlds of Genetics

The most important development during this century of behavioral genetic research has been the synthesis of the two worlds of genetics, quantitative genetics and molecular genetics. Quantitative genetics and molecular genetics both have their origins in the 1860s with Francis Galton (Galton 1865 , 1869 ) and Gregor Mendel (Mendel 1866 ), respectively. Not much happened until the 1900s when Galton’s insights led to methods to study genetic influence on complex traits and when Mendel’s work was re-discovered. The two worlds clashed as Mendelians looked for 3:1 segregation ratios indicative of single-gene traits, whereas Galtonians assumed that Mendel’s laws of heredity were specific to pea plants because they knew that complex traits are distributed continuously.

Antipathy between the two worlds of genetics followed because of the different goals of Mendelians and Galtonians. Mendelians, the predecessors of molecular geneticists, wanted to understand how genes work, which led to the use of induced mutations and a focus on dichotomous traits that were easily assessed such as physical characteristics rather than behavioral traits. In contrast, Galtonians, whose descendants are quantitative geneticists, used genetics as a tool to understand the etiology of naturally occurring variation in complex traits selected for their intrinsic interest and importance, with behavioral traits, especially intelligence, high on the list. The resolution to the conflict could be seen in Ronald Fisher’s 1918 paper, which showed that Mendelian inheritance is compatible with quantitative traits if the assumption is made that several genes affect a trait (Fisher 1918 ). Nonetheless, the two worlds of genetics went their own way for most of the century.

The synthesis of the two worlds of genetics began in the 1980s with the technological advances of DNA sequencing, polymerase chain reaction, and DNA microarrays that enabled genome-wide association (GWA) studies of complex traits. In addition to finding DNA variants associated with complex traits, GWA genotypes led to three far-reaching advances in genetic research. First, GWA genotypes were used to estimate directly the classical quantitative genetic parameters of heritability and genetic correlation, which could be called quantitative genomics . Second, the results of GWA studies were used to create polygenic scores that predict individual differences for complex traits. Third, GWA genotypes facilitated new approaches to causal modeling of the interplay between genes and environment. Together, when applied to behavioral traits, these advances could be called behavioral genomics . This synthesis of the two worlds of genetics, the journey from behavioral genetics to behavioral genomics, is the overarching theme of this whistle-stop tour celebrating a century of research in behavioral genetics. (See Fig.  1 .) The itinerary begins with milestones in quantitative genetics and then molecular genetics, concluding with behavioral genomics.

Synthesis of the two worlds of genetics: from behavioral genetics to behavioral genomics.

Quantitative Genetics

The first 50 years of quantitative genetic research, from 1920 to 1970, started off well with family studies (Jones 1928 ; Thorndike 1928 ), twin studies (Holzinger 1929 ; Lauterbach 1925 ; Merriman 1924 ; Tallman 1928 ) and adoption studies (Burks 1928 ; Freeman et al. 1928 ) using the recently devised IQ test. However, this nascent research was squelched with the emergence of Nazi eugenic policies (McGue 2008 ). The void was filled with behaviorism (Watson 1930 ), which led to environmentalism, the ‘blank slate’ view that we are what we learn (Pinker 2003 ).

Nonetheless, a few studies of IQ appeared in the 1930 and 1940 s, such as the first study of identical twins reared apart (Newman et al. 1937 ) and the first adoption study that assessed birth parents (Skodak and Skeels 1949 ). Both indicated substantial genetic influence on IQ, as did a review of all available IQ data (Woodworth 1941 ).

In 1960, the field-defining book, Behavior Genetics (Fuller and Thompson 1960 ), was published. It mostly reviewed research on nonhuman animals. In their preface, the authors noted that “we considered omitting human studies completely” (p. vi); even their chapter on cognitive abilities primarily reviewed nonhuman research. An earlier influential review began by saying, “In the writer’s opinion, the genetics of behavior must be worked out on species that can be subjected to controlled breeding. At the present time this precludes human subjects” (Hall 1951 ).

In 1963, a milestone review was published in Science of 52 family, twin and adoption studies of IQ (Erlenmeyer-Kimling and Jarvik 1963 ). Although the studies were very small by modern standards and heritability was not calculated, the average results from the different designs suggested substantial heritability. For example, the average MZ and DZ twin correlations were 0.87 and 0.53, respectively, suggesting a heritability of 68%. However, despite being published in Science , the paper was largely ignored; it was cited only 22 times in five years.

The pace of behavioral genetic research picked up in the 1960s, once again primarily research on non-human animals (Lindzey et al. 1971 ; McClearn 1971 ), although some twin studies on cognitive abilities were also published (Nichols 1965 ; Schoenfeldt 1968 ). However, the first 50 years of quantitative genetic research ended badly with the publication in 1969 of Arthur Jensen’s paper, How Much Can We Boost IQ and Scholastic Achievement? (Jensen 1969 ). The paper touched on ethnic differences, which made it one of the most controversial papers in the behavioral sciences, with 900 citations in the first five years and more than 6200 citations in total.

1970 was a watershed year marking the second 50 years of behavioral genetic research. It was the year that the Behavior Genetics Association was launched and the first issue of its journal, Behavior Genetics , was published. Another 1970 milestone was the publication of the foundational paper for model-fitting analysis of quantitative genetic designs (Jinks and Fulker 1970 ).

The 1970s and 1980s yielded most of the major discoveries for quantitative genetics as applied to behavioral traits, discoveries that are listed as landmarks in the following paragraphs. Nonetheless, in the aftermath of Jensen’s 1969 paper, behavioral genetic research, especially on intelligence, was highly controversial (Scarr and Carter-Saltzman 1982 ). Most notably, Leon Kamin severely criticized the politics as well as science of behavioral genetic research on intelligence in his book, The Science and Politics of I.Q. (Kamin 1974 ). He concluded that “There exist no data which should lead a prudent man to accept the hypothesis that I.Q. test scores are in any degree heritable” (p. 1). The book was cited more than 2000 times and stoked antipathy towards genetic research. It also impugned the motivation of genetic researchers, saying that they are ‘committed to the view that those on the bottom are genetically inferior victims of their own immutable defects’ (p. 2).

All Traits are Heritable

Despite this hostility, genetic research grew exponentially in the 1970s and created a seismic shift from the prevailing view that behavioral traits like intelligence are not “in any degree heritable”. In 1978, a review of 30 twin studies of intelligence yielded an average heritability estimate of 46% (Nichols 1978 ). Moreover, the conclusion began to emerge that all traits show substantial heritability. This conclusion, which has been called the first law of behavioral genetics (Turkheimer 2000 ), was first observed in 1976 in a twin study of cognitive data for 3000 twin pairs, which also included extensive data on personality and interests for 850 twin pairs (Loehlin and Nichols 1976 ). The authors noted “the curious uniformity of identical-fraternal differences both within and across trait domains” (p. 89). A 2015 meta-analysis of all published twin studies showed that behavioral traits are about 50% heritable on average (Polderman et al. 2015 ). Demonstrating the ubiquitous importance of genetics was the fundamental accomplishment of behavioral genetics.

No Traits are 100% Heritable

The flip side of the finding of 50% heritability was just as important: no traits are 100% heritable. It is ironic that, after a century of environmentalism, genetic research provided the strongest evidence for the importance of the environment; previous environmental research was confounded because it ignored genetics. Moreover, investigating environmental influences in genetically sensitive designs led to two of the most important discoveries about the environment: nonshared environment and the nature of nurture.

Nonshared Environment

Quantitative genetic research showed that environmental influences work very differently from the way they were assumed to work. A second discovery by Loehlin and Nichols ( 1976 ) was that salient environmental influences are not shared by twins growing up in the same family: “Environment carries substantial weight in determining personality – it appears to account for at least half the variance – but that environment is one for which twin pairs are correlated close to zero” (p. 92). This phenomenon has come to be known as nonshared environment (Plomin and Daniels 1987 ).

Loehlin and Nichols suggested that cognitive abilities are an exception to the rule that environmental influences make children in a family different from, not similar to, one another. Their twin study suggested that about 25% of the variance of cognitive abilities could be attributed to shared environment. A direct test of shared environmental influence is the correlation between adoptive siblings, genetically unrelated children adopted into the same family. Seven small studies of adoptive siblings yielded an average IQ correlation of 0.25, which seemed to precisely confirm the twin estimate (McGue et al. 1993 ).

However, in 1978, a study of 100 pairs of adoptive siblings reported an IQ correlation of -0.03 (Scarr and Weinberg 1978 ). This is a good example of the progressive nature of behavioral genetic research (Urbach 1974 ). Scarr and Weinberg noted that previous studies involved children, whereas theirs was the first study of post-adolescent adoptive siblings aged 16 to 22, and they hypothesized that the effect of shared environmental influence on cognitive development diminishes after adolescence as young adults make their own way in the world. Their hypothesis was confirmed in two additional studies of post-adolescent adoptive siblings that yielded an average IQ correlation of -0.01 (McGue et al. 1993 ). Evidence that shared environmental influence declines after adolescence to negligible levels for cognitive abilities has also emerged from twin studies (Briley and Tucker-Drob 2013 ; Haworth et al. 2010 ). However, one of the biggest mysteries about nonshared environment remains: what are these environmental influences that make children growing up in the same family so different (Plomin 2011 )?

The Nature of Nurture

Another milestone was the revelation that environmental measures widely used in the behavioral sciences, such as parenting, social support, and life events, show genetic influence (Plomin and Bergeman 1991 ), with heritabilities of about 25% on average (Kendler and Baker 2007 ). This finding emerged in the 1980s as measures of the environment were included in quantitative genetic designs, which also led to the discovery that associations between environmental measures and psychological traits are significantly mediated genetically (Plomin et al. 1985 ). The nature of nurture is one of the major directions for research in behavioral genomics, as discussed later.

Heritability Increases During Development

Another milestone in the 1970s was the Louisville Twin Study in which mental development of 500 pairs of twins was assessed longitudinally and showed that the heritability of intelligence increases from infancy to adolescence (Wilson 1983 ). In light of the replication crisis in science (Ritchie 2021 ), a cause for celebration is that this counterintuitive finding of increasing heritability of intelligence – from about 40% in childhood to more than 60% in adulthood -- has consistently replicated, as seen in cross-sectional (Haworth et al. 2010 ) and longitudinal (Briley and Tucker-Drob 2013 ) mega-analyses.

In 1977, a landmark paper showed how univariate analysis of variance can be extended to multivariate analysis of covariance in a model-fitting framework (Martin and Eaves 1977 ). They applied their approach to cognitive abilities and found an average genetic correlation of 0.52, indicating that many genes affect diverse traits, called pleiotropy . Subsequent studies also yielded genetic correlations greater than 0.50 between diverse cognitive abilities (Plomin and Kovas 2005 ).

In the 1970s and 1980s, bigger and better studies made most of the major quantitative genetic discoveries, going far beyond merely estimating heritability. But it was not all smooth sailing. Most notably, The Bell Curve resurrected many of the issues that followed Jensen’s 1969 paper (Herrnstein and Murray 1996 ). Nonetheless, by the 1990s, quantitative genetic research had convinced most scientists of the importance of genetics for behavioral traits, including intelligence (Snyderman and Rothman 1990 ). One symbol of this change was that the 1992 Centennial Conference of the American Psychological Association chose behavioral genetics as one of two themes that best represented the past, present, and future of psychology (Plomin and McClearn 1993 ). Then, just as quantitative genetic discoveries began to slow, the synthesis with molecular genetics began, which led to the DNA revolution and behavioral genomics.

Molecular Genetics

During its first 50 years, molecular genetics focused on single-gene disorders. In 1933, a Nobel prize was awarded to Thomas Hunt Morgan for mapping genes responsible for single-gene mutations in fruit flies (Morgan et al. 1923 ), but human mapping was stymied because only a few single-gene markers such as blood types were available – variants in DNA itself were not available for another fifty years. Research on single-gene effects discovered in pedigree studies only incidentally involved behavioral traits. For example, phenylketonuria, the most common single-gene metabolic disorder, was discovered in 1934 (Folling 1934 ) and shown to be responsible for 1% of the population institutionalized for severe intellectual disability.

In the 1940s, it became clear that DNA is the mechanism of heredity, culminating in the most famous paper in biology which proposed the double-helix structure of DNA (Watson and Crick 1953 ). An important milestone for human behavioral genetics was the discovery in 1959 that the most common form of intellectual disability, Down syndrome, was due to a trisomy of chromosome 21 (Lejeune et al. 1959 ).

In 1961, the genetic code was cracked showing that three-letter sequences of the four-letter alphabet of DNA coded for the 20 amino acids (Crick et al. 1961 ). Just as with quantitative genetics, the 1970s was a watershed decade that ushered in the second 50 years, the genomics era.

The Genomics Era

The era of genomics began in the 1970s when methods were developed to sequence DNA’s nucleotide bases (Sanger et al. 1977 ). In 2003, fifty years after the discovery of the double helix structure of DNA, the Human Genome Project identified the sequence of 92% of the three billion nucleotide bases in the human genome (Collins et al. 2003 ).

In the 1980s, the first common variants in DNA itself were discovered, restriction fragment length polymorphisms (RFLPs) (Botstein et al. 1980 ). RFLPs enabled linkage mapping for single-gene disorders and were the basis for DNA fingerprinting, which revolutionized forensics (Jeffreys 1987 ). Polymerase chain reaction (PCR) was also developed which facilitated genotyping by rapidly amplifying DNA fragments (Mullis et al. 1986 ). In the 1980s, these developments increased the pace of linkage mapping of single-gene disorders, many of which had cognitive consequences, such as phenylketonuria (Woo et al. 1983 ) and Huntington disease (Gusella et al. 1983 ). In the 1990s, DNA sequencing revealed thousands of single-nucleotide polymorphisms (SNPs), the most common DNA variant (Collins et al. 1997 ).

In the 1990s, linkage was also attempted for complex traits that did not show single-gene patterns of transmission, such as reading disability (Cardon et al. 1994 ), but these were unsuccessful because linkage, which traces chromosomal recombination between disease genes and DNA variants within families, is unable to detect small effect sizes (Plomin et al. 1994 ). Researchers then pivoted towards allelic association in unrelated individuals, which is much more powerful in detecting DNA variants of small effect size. An early example of association was an allele of the apolipoprotein E gene on chromosome 19 that was found in 40% of individuals with late-onset Alzheimer disease as compared to 15% in controls (Corder et al. 1993 ).

The downside of allelic association is that an association can only be detected if a DNA variant is itself the functional gene or very close to it. For this reason, and because genotyping each DNA variant was slow and expensive, the 1990s became the decade of candidate gene studies in which thousands of studies reported associations between complex behavioral traits and a few ‘candidate’ genes, typically neurotransmitter genes thought to be involved in behavioral pathways. However, these candidate-gene associations failed to replicate because these studies committed most of the sins responsible for the replication crisis (Ioannidis 2005 ). For example, when 12 candidate genes reported to be associated with intelligence were tested in three large samples, none replicated (Chabris et al. 2012 ).

Genome-wide Association

In 1996, an idea emerged that was the opposite of the candidate-gene approach: using thousands of DNA variants to systematically assess associations across the genome in large samples of unrelated individuals (Risch and Merikangas 1996 ). However, genome-wide association (GWA) seemed a dream because genotyping was slow and expensive.

The problem of genotyping each DNA variant in large samples was solved in the 2000s by the commercial availability of DNA microarrays, called SNP chips , which genotype hundreds of thousands of SNPs for an individual quickly, accurately, and inexpensively. SNP chips paved the way for GWA analyses. In 2007, the first major GWA analysis included 2000 cases for each of seven major disorders and compared SNP allele frequencies for these cases with controls (The Wellcome Trust Case Control Consortium 2007 ). Replicable associations were found but they were few in number and extremely small in effect size. Hundreds of GWA reports appeared in the next decade with similarly small effect sizes across the behavioral and biological sciences (Visscher et al. 2017 ), including cognitive traits such as educational attainment (Rietveld et al. 2013 ) and intelligence in childhood (Benyamin et al. 2014 ) and adulthood (Davies et al. 2011 ).

These GWA studies led to the realization that the biggest effect sizes were much smaller than anyone anticipated. For case-control studies, risk ratios were less than 1.1, and for dimensional traits, variance explained was less than 0.001. This meant that huge sample sizes would be needed to detect these miniscule effects, and thousands of these associations would be needed to account for heritability, which is usually greater than 50% for cognitive traits. Ever larger GWA samples scooped up more of these tiny effects. Most recently, a GWA meta-analysis with a sample size of 3 million netted nearly four thousand independent significant associations after correction for multiple testing, but the median effect size of these SNPs accounted for less than 0.0001 of the variance (Okbay et al. 2022 ).

A century after Fisher’s 1918 paper, the discovery of such extreme polygenicity (Boyle et al. 2017 ; Visscher et al. 2021 ) was a turning point in the voyage from behavioral genetics to behavioral genomics. GWA genotypes brought the two worlds of genetics together by making it possible to use GWA genotypes to create three sets of tools to investigate highly polygenic traits: quantitative genomics, polygenic scores, and causal modeling (see Fig.  1 ). When applied to behavioral traits, these tools constitute the new field of behavioral genomics.

Quantitative Genomics

What good are SNP associations that account for such tiny effects? The molecular genetic goal of tracking effects from genes to brain to behavior is daunting when the effects are so small. However, in contrast to this bottom-up approach from genes to behavior, the top-down perspective of behavioral genetics answered this question by using GWA genotypes to estimate quantitative genetic parameters of heritability and genetic correlations, which could be called quantitative genomics . The journey picked up speed as quantitative genomics led to three new milestones.

Genome-wide Complex Trait Analysis (GCTA). In 2011, the first new method was devised to estimate heritability and genetic correlations since twin and adoption designs in the early 1900s. GCTA (originally called GREML) uses GWA genotypes for large samples of unrelated individuals to compare overall SNP similarity to phenotypic similarity pair by pair for all pairs of individuals (Yang et al. 2011 ). The extent to which SNP similarity explains trait similarity is called SNP heritability because it is limited to heritability estimated by the SNPs on the SNP chip. Genetic correlations are estimated by comparing each pair’s SNP similarity to their cross-trait phenotypic similarity.

SNP heritability estimates are about half the heritability estimated by twin studies (Plomin and von Stumm 2018 ). This ‘missing heritability’ occurs because SNP heritability is limited to the common SNPs genotyped on current SNP chips, which also creates a ceiling for discovery in GWA research. Most SNPs are not common, and rare SNPs appear to be responsible for much of the missing heritability, at least for height (Wainschtein et al. 2022 ). Importantly, quantitative genomic estimates of genetic correlations are not limited in this way and thus provide estimates of genetic correlations similar to those from twin studies (Trzaskowski et al. 2013 ).

Linkage Disequilibrium Score (LDSC) Regression. In 2015, a second quantitative genomic method, LDSC, was published which estimates heritability and genetic correlations from GWA summary effect size statistics for each SNP, corrected for linkage disequilibrium between SNPs (Bulik-Sullivan et al. 2015 ). LDSC estimates of heritability and genetic correlations are similar to GCTA estimates, although GCTA estimates are generally more accurate (Evans et al. 2018 ; Ni et al. 2018 ). The advantage of LDSC is that it can be applied to published GWA summary statistics in contrast to GCTA which requires access to GWA data for individuals in the GWA study.

Genomic Structural Equation Modeling (Genomic SEM). In 2019, a third quantitative genomic analysis completed the arc from quantitative genetics to quantitative genomics by combining quantitative genetic structural equation model-fitting, routinely used in twin analyses, to LDSC heritabilities and genetic correlations (Grotzinger et al. 2019 ). Genomic SEM provides insights into the multivariate genetic architecture of cognitive traits (Grotzinger et al. 2019 ) and psychopathology (Grotzinger et al. 2022 ).

The second answer to the question about what to do with SNP associations that have such small effect sizes is the creation of polygenic scores.

Polygenic Scores

A milestone that marks the spot where the DNA revolution began to transform the behavioral sciences is polygenic scores. Rather than using GWA genotypes to estimate SNP heritabilities and genetic correlations, polygenic scores use GWA genotypes to create a single score for each individual that aggregates, across all SNPs on a SNP chip, an individual’s genotype for each SNP (0, 1 or 2) weighted by the SNP’s effect size on the target trait as indicated by GWA summary statistics. In 2001, polygenic scores were introduced in plant and animal breeding (Meuwissen et al. 2001 ) and later in cognitive abilities (Harlaar et al. 2005 ) and psychopathology (Purcell et al. 2009 ). GWA summary statistics needed to create polygenic scores are now publicly available for more than 500 traits, including dozens for psychiatric disorders and other behavioral traits including cognitive traits (PGS Catalog 2022 ).

The most predictive polygenic scores in the behavioral sciences are for cognitive traits, especially educational attainment and intelligence. Early GWA studies of cognitive traits were underpowered to detect the small effects that we now know are responsible for heritability (Plomin and von Stumm 2018 ). In 2013, a landmark was a GWA study of educational attainment with a sample size exceeding 100,000 (Rietveld et al. 2013 ). A polygenic score derived from its GWA summary statistics predicted 2% of the variance of educational attainment in independent samples. The finding that the biggest effects accounted for only 0.0002 of the variance of educational attainment made it clear that much larger samples would be needed to scoop up more of the tiny effects responsible for the twin heritability estimate of about 40%. In the past decade, the predictive power of polygenic scores for educational attainment has increased with increasing sample sizes from 2% (Rietveld et al. 2013 ) to 5% (Okbay et al. 2016 ) to 10% (Lee et al. 2018 ) to 14% in a GWA study with a sample size of three million (Okbay et al. 2022 ). The current polygenic score for intelligence, derived from a GWA study with a sample of 280,000, predicted 4% of the variance (Savage et al. 2018 ), but, together, the polygenic scores for educational attainment and intelligence predicted 10% of the variance of intelligence test scores (Allegrini et al. 2019 ).

The next milestone will be to narrow the gap between heritability explained by polygenic scores and SNP heritability. A more daunting challenge will be to break through the ceiling of SNP heritability to reach the heritability estimated by twin studies. Reaching both of these destinations will be facilitated by even larger GWA studies and whole-genome sequencing (Wainschtein et al. 2022 ).

Polygenic scores are unique predictors because inherited DNA variations do not change systematically during life – there is no backward causation in the sense that nothing in the brain, behavior or environment changes inherited differences in DNA sequence. For this reason, polygenic scores can predict behavioral traits from early in life without knowing anything about the intervening pathways between genes, brain, and behavior.

Polygenic scores have brought behavioral genetics to the forefront of research in many areas of the life sciences because polygenic scores can be created in any sample of unrelated individuals for whom GWA genotype data are available. No special samples of twins or adoptees are needed, nor is it necessary to assess behavioral traits in order to use polygenic scores to predict them.

Although the implications and applications of polygenic scores derive from its power to predict behavioral traits without regard to explanation (Plomin and von Stumm 2022 ), another milestone on the road to behavioral genomics has been the leverage provided by GWA genotypes for causal modeling.

Causal Modeling

A final milestone on the journey from behavioral genetics to behavioral genomics is a suite of new approaches that use GWA genotypes in causal models that attempt to dissect sources of genetic influence on behavioral traits (Pingault et al. 2018 ). Although traditional quantitative genetic models are causal models, GWA genotypes have enhanced causal modeling in research on assortative mating (Border et al. 2021 ; Yengo et al. 2018 ), population stratification (Abdellaoui et al. 2022 ; Lawson et al. 2020 ), and Mendelian randomization (Richmond and Davey Smith 2022 ).

An explosion of research on genotype-environment correlation was ignited by a 2018 paper in Science on the topic of the nature of nurture (Kong et al. 2018 ). The study included both parent and offspring GWA genotypes and showed that a polygenic score computed from non-transmitted alleles from parent to offspring influenced offspring educational attainment; these indirect effects were dubbed genetic nurture . GCTA has also been used to investigate genotype-environment correlation (Eilertsen et al. 2021 ). Although a great strength of behavioral genomics is its ability to investigate genetic influence in samples of unrelated individuals, combining GWA genotypes with traditional quantitative genetic designs has also enriched causal modeling (McAdams et al. 2022 ), for example, by comparing results within and between families (Brumpton et al. 2020 ; Howe et al. 2022 ).

This whistle-stop tour has highlighted some of the milestones in a century of research in behavioral genetics. The progress is unmatched in the behavioral sciences and its discoveries have been transformative. The most exciting development is the synthesis of quantitative genetics and molecular genetics into behavioral genomics. The energy from this fusion will propel the field far into the future.

Acknowledgements

This work was supported in part by the UK Medical Research Council (MR/V012878/1 and previously MR/M021475/1).

Author Contributions

The author wrote, reviewed, and submitted the manuscript.

Declarations

The author has no relevant financial or non-financial interests to disclose.

This review was based on a talk given at the 52nd Behavior Genetics Association Annual Meeting, Los Angeles, California, June 25, 2022.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

• Abdellaoui A, Dolan CV, Verweij KJH, Nivard MG. Gene–environment correlations across geographic regions affect genome-wide association studies. Nat Genet. 2022; 54 :1345–1354. doi: 10.1038/s41588-022-01158-0. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Allegrini AG, Selzam S, Rimfeld K, et al. Genomic prediction of cognitive traits in childhood and adolescence. Mol Psychiatry. 2019; 24 :819–827. doi: 10.1038/s41380-019-0394-4. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Benyamin B, Pourcain Bs, Davis OS, et al. Childhood intelligence is heritable, highly polygenic and associated with FNBP1L. Mol Psychiatry. 2014; 19 :253–258. doi: 10.1038/mp.2012.184. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Border R, O’Rourke S, de Candia T et al (2021) Assortative mating biases marker-based heritability estimators. Nat Genet 13:660. 10.1038/s41467-022-28294-9 [ PMC free article ] [ PubMed ]
• Botstein D, White RL, Skolnick M, Davis RW. Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am J Hum Genet. 1980; 32 :314–331. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Boyle EA, Li YI, Pritchard JK. An expanded view of complex traits: from polygenic to omnigenic. Cell. 2017; 169 :1177–1186. doi: 10.1016/j.cell.2017.05.038. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Briley DA, Tucker-Drob EM. Explaining the increasing heritability of cognitive ability across development: a meta-analysis of longitudinal twin and adoption studies. Psychol Sci. 2013; 24 :1704–1713. doi: 10.1177/0956797613478618. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Brumpton B, Sanderson E, Heilbron K, et al. Avoiding dynastic, assortative mating, and population stratification biases in mendelian randomization through within-family analyses. Nat Commun. 2020; 11 :3519. doi: 10.1038/s41467-020-17117-4. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bulik-Sullivan B, ReproGen C, Psychiatric Genomics Consortium et al. An atlas of genetic correlations across human diseases and traits. Nat Genet. 2015; 47 :1236–1241. doi: 10.1038/ng.3406. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Burks BS. The relative influence of nature and nurture upon mental development; a comparative study of foster parent-foster child resemblance and true parent-true child resemblance. Teachers Coll Record. 1928; 29 :219–316. doi: 10.1177/016146812802900917. [ CrossRef ] [ Google Scholar ]
• Cardon LR, Smith SD, Fulker DW, et al. Quantitative trait locus for reading disability on chromosome 6. Science. 1994; 266 :276–279. doi: 10.1126/science.7939663. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Chabris CF, Hebert BM, Benjamin DJ, et al. Most reported genetic associations with general intelligence are probably false positives. Psychol Sci. 2012; 23 :1314–1323. doi: 10.1177/0956797611435528. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Collins FS, Green ED, Guttmacher AE, Guyer MS. A vision for the future of genomics research. Nature. 2003; 422 :835–847. doi: 10.1038/nature01626. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Collins FS, Guyer MS, Chakravarti A. Variations on a theme: cataloging human dna sequence variation. Science. 1997; 278 :1580–1581. doi: 10.1126/science.278.5343.1580. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Corder EH, Saunders AM, Strittmatter WJ, et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science. 1993; 261 :921–923. doi: 10.1126/science.8346443. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Crick FHC, Barnett L, Brenner S, Watts-Tobin RJ. General nature of the genetic code for proteins. Nature. 1961; 192 :1227–1232. doi: 10.1038/1921227a0. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Davies G, Tenesa A, Payton A, et al. Genome-wide association studies establish that human intelligence is highly heritable and polygenic. Mol Psychiatry. 2011; 16 :996–1005. doi: 10.1038/mp.2011.85. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Eilertsen EM, Jami ES, McAdams TA, et al. Direct and indirect effects of maternal, paternal, and offspring genotypes: Trio-GCTA. Behav Genet. 2021; 51 :154–161. doi: 10.1007/s10519-020-10036-6. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Erlenmeyer-Kimling L, Jarvik LF. Genetics and intelligence: a review. Science. 1963; 142 :1477–1479. doi: 10.1126/science.142.3598.1477. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Evans LM, Tahmasbi R, Vrieze SI, et al. Comparison of methods that use whole genome data to estimate the heritability and genetic architecture of complex traits. Nat Genet. 2018; 50 :737–745. doi: 10.1038/s41588-018-0108-x. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Fisher RA. The correlation between relatives on the supposition of mendelian inheritance. Trans R Soc Edinb. 1918; 52 :399–433. doi: 10.1017/S0080456800012163. [ CrossRef ] [ Google Scholar ]
• Folling A (1934) Excretion of phenylpyruvic acid in urine as a metabolicnomaly in connection with imbecility.Nord Med Tidskr 8:1054-1059
• Freeman FN, Holzinger KJ, Mithell BC. The influence of environment on the intelligence, school achievement, and conduct of foster children. Yearbook of the National Society for the Study of Education. 1928; 1 :102–217. [ Google Scholar ]
• Fuller JL, Thompson WR. Behavior genetics. New York: Wiley; 1960. [ Google Scholar ]
• Galton F. Hereditary talent and character. Macmillan’s Magazine. 1865; 12 :157–166. [ Google Scholar ]
• Galton F. Hereditary genius: an inquiry into its laws and consequences. London: Collins; 1869. [ Google Scholar ]
• Grotzinger AD, Mallard TT, Akingbuwa WA, et al. Genetic architecture of 11 major psychiatric disorders at biobehavioral, functional genomic and molecular genetic levels of analysis. Nat Genet. 2022; 54 :548–559. doi: 10.1038/s41588-022-01057-4. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Grotzinger AD, Rhemtulla M, de Vlaming R, et al. Genomic structural equation modelling provides insights into the multivariate genetic architecture of complex traits. Nat Hum Behav. 2019; 3 :513–525. doi: 10.1038/s41562-019-0566-x. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Gusella JF, Wexler NS, Conneally PM, et al. A polymorphic DNA marker genetically linked to Huntington’s disease. Nature. 1983; 306 :234–238. doi: 10.1038/306234a0. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hall CS. The genetics of behavior. In: Stevens SS, editor. Handbook of experimental psychology. New York: Wiley; 1951. pp. 304–329. [ Google Scholar ]
• Harlaar N, Butcher LM, Meaburn E, et al. A behavioural genomic analysis of DNA markers associated with general cognitive ability in 7-year-olds. J Child Psychol & Psychiat. 2005; 46 :1097–1107. doi: 10.1111/j.1469-7610.2005.01515.x. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Haworth CMA, Wright MJ, Luciano M, et al. The heritability of general cognitive ability increases linearly from childhood to young adulthood. Mol Psychiatry. 2010; 15 :1112–1120. doi: 10.1038/mp.2009.55. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Herrnstein RJ, Murray CA. The bell curve: intelligence and class structure in american life, 1st Free Press pbk. New York: ed. Simon & Schuster; 1996. [ Google Scholar ]
• Holzinger KJ. The relative effect of nature and nurture influences on twin differences. J Educ Psychol. 1929; 20 :241–248. doi: 10.1037/h0072484. [ CrossRef ] [ Google Scholar ]
• Howe LJ, Nivard MG, Morris TT, et al. Within-sibship genome-wide association analyses decrease bias in estimates of direct genetic effects. Nat Genet. 2022; 54 :581–592. doi: 10.1038/s41588-022-01062-7. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ioannidis JPA. Why most published research findings are false. PLoS Med. 2005; 2 :e124. doi: 10.1371/journal.pmed.0020124. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jeffreys AJ. Highly variable minisatellites and DNA fingerprints. Biochem Soc Trans. 1987; 15 :309–317. doi: 10.1042/bst0150309. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jensen A. How much can we boost iq and scholastic achievement. Harv Educational Rev. 1969; 39 :1–123. doi: 10.17763/haer.39.1.l3u15956627424k7. [ CrossRef ] [ Google Scholar ]
• Jinks JL, Fulker DW. Comparison of the biometrical genetical, MAVA, and classical approaches to the analysis of the human behavior. Psychol Bull. 1970; 73 :311–349. doi: 10.1037/h0029135. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jones HE. A first study of parent-child resemblance in intelligence. Yearbook of the National Society for the Study of Education Part. 1928; 1 :61–72. [ Google Scholar ]
• Kamin LJ. The science and politics of I.Q. New York, NY: Routledge; 1974. [ PubMed ] [ Google Scholar ]
• Kendler KS, Baker JH. Genetic influences on measures of the environment: a systematic review. Psychol Med. 2007; 37 :615–626. doi: 10.1017/S0033291706009524. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kong A, Thorleifsson G, Frigge ML, et al. The nature of nurture: Effects of parental genotypes. Science. 2018; 359 :424–428. doi: 10.1126/science.aan6877. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lauterbach CE. Studies in twin resemblance. Genetics. 1925; 10 :525–568. doi: 10.1093/genetics/10.6.525. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lawson DJ, Davies NM, Haworth S, et al. Is population structure in the genetic biobank era irrelevant, a challenge, or an opportunity? Hum Genet. 2020; 139 :23–41. doi: 10.1007/s00439-019-02014-8. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lee JJ, Wedow R, Okbay A, et al. Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals. Nat Genet. 2018; 50 :1112–1121. doi: 10.1038/s41588-018-0147-3. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lejeune J, Gautier M, Turpin R. [Study of somatic chromosomes from 9 mongoloid children] C R Hebd Seances Acad Sci. 1959; 248 :1721–1722. [ PubMed ] [ Google Scholar ]
• Lindzey G, Loehlin J, Manosevitz M, Thiessen D. Behavioral Genetics. Annu Rev Psychol. 1971; 22 :39–94. doi: 10.1146/annurev.ps.22.020171.000351. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Loehlin JC. History of behavior genetics. In: Kim Y-K, editor. Handbook of behavior genetics. New York, NY: Springer New York; 2009. pp. 3–11. [ Google Scholar ]
• Loehlin JC, Nichols RC. Heredity, environment, and personality: a study of 850 sets of twins. Austin, Texas: University of Texas Press; 1976. [ Google Scholar ]
• Martin NG, Eaves LJ. The genetical analysis of covariance structure. Heredity. 1977; 38 :79–95. doi: 10.1038/hdy.1977.9. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Maxson SC (2007) A history of behavior genetics. In: Jones BC, Mormee P (eds) Neurobehavioral genetics: methods and applications, 2nd edn. CRC Press, New York, pp 1–16
• McAdams TA, Cheesman R, Ahmadzadeh YI (2022) Annual Research Review: towards a deeper understanding of nature and nurture: combining family-based quasi‐experimental methods with genomic data. Child Psychol Psychiatry jcpp 13720. 10.1111/jcpp.13720 [ PMC free article ] [ PubMed ]
• McClearn GE. Behavioral genetics. Syst Res. 1971; 16 :64–81. doi: 10.1002/bs.3830160106. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• McGue M. The end of behavioral genetics? Acta Physiol Sinica. 2008; 40 :1073–1087. doi: 10.3724/sp.j.1041.2008.01073. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• McGue M, Bouchard TJ, Iacono WG, Lykken . Nature, nurture and psychology. Washington, DC: American Psychological Association; 1993. Behavioral genetics of cognitive ability: a life-span perspective; pp. 59–76. [ Google Scholar ]
• Mendel G. Versuche ueber Pflanzenhybriden. Verhandllungen des Naturforschhunden Vereines in Bruenn. 1866; 4 :3–47. [ Google Scholar ]
• Merriman C. The intellectual resemblance of twins. Psychol Monogr. 1924; 33 :i–57. doi: 10.1037/h0093212. [ CrossRef ] [ Google Scholar ]
• Meuwissen THE, Hayes BJ, Goddard ME. Prediction of total genetic value using genome-wide dense marker maps. Genetics. 2001; 157 :1819–1829. doi: 10.1093/genetics/157.4.1819. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Morgan TH, Sturtevant AH, Muller HJ, Bridges CB (1923) The mechanism of Mendelian heredity. H. Holt and Company
• Mullis K, Faloona F, Scharf S, et al. Specific enzymatic amplification of dna in vitro: the polymerase chain reaction. Cold Spring Harb Symp Quant Biol. 1986; 51 :263–273. doi: 10.1101/SQB.1986.051.01.032. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Newman HH, Freeman FN, Holzinger KJ (1937) Twins: a study of heredity and environment. University of Chicago Press
• Ni G, Moser G, Wray NR, et al. Estimation of genetic correlation via linkage disequilibrium score regression and genomic restricted maximum likelihood. Am J Hum Genet. 2018; 102 :1185–1194. doi: 10.1016/j.ajhg.2018.03.021. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Nichols RC. The National Merit Twin Study. In: Vandenberg SG, editor. Methods and goals of human behavior genetics. New York: Academic Press; 1965. pp. 231–243. [ Google Scholar ]
• Nichols RC. Twin studies of ability, personality, and interests. Homo. 1978; 29 :158–173. [ Google Scholar ]
• Okbay A, Beauchamp JP, Fontana MA, et al. Genome-wide association study identifies 74 loci associated with educational attainment. Nature. 2016; 533 :539–542. doi: 10.1038/nature17671. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Okbay A, Wu Y, Wang N, et al. Polygenic prediction of educational attainment within and between families from genome-wide association analyses in 3 million individuals. Nat Genet. 2022; 54 :437–449. doi: 10.1038/s41588-022-01016-z. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• PGS Catalog (2022) PGS Catalog. http://www.pgscatalog.org/
• Pingault J-B, O’Reilly PF, Schoeler T, et al. Using genetic data to strengthen causal inference in observational research. Nat Rev Genet. 2018; 19 :566–580. doi: 10.1038/s41576-018-0020-3. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pinker S. The blank slate: the modern denial of human nature. London: Nachdr. Penguin; 2003. [ Google Scholar ]
• Plomin R. Commentary: why are children in the same family so different? Non-shared environment three decades later. Int J Epidemiol. 2011; 40 :582–592. doi: 10.1093/ije/dyq144. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Plomin R, Bergeman CS. The nature of nurture: genetic influence on “environmental” measures. Behav Brain Sci. 1991; 14 :373–386. doi: 10.1017/S0140525X00070278. [ CrossRef ] [ Google Scholar ]
• Plomin R, Daniels D. Why are children in the same family so different from one another? Behav Brain Sci. 1987; 10 :1–16. doi: 10.1017/S0140525X00055941. [ CrossRef ] [ Google Scholar ]
• Plomin R, Kovas Y. Generalist genes and learning disabilities. Psychol Bull. 2005; 131 :592–617. doi: 10.1037/0033-2909.131.4.592. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Plomin R, Loehlin JC, DeFries JC. Genetic and environmental components of “environmental” influences. Dev Psychol. 1985; 21 :391–402. doi: 10.1037/0012-1649.21.3.391. [ CrossRef ] [ Google Scholar ]
• Plomin R, McClearn GE (eds) (1993) Nature, nurture, & psychology. American Psychological Association, Washington, DC
• Plomin R, Owen M, McGuffin P. The genetic basis of complex human behaviors. Science. 1994; 264 :1733–1739. doi: 10.1126/science.8209254. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Plomin R, von Stumm S. The new genetics of intelligence. Nat Rev Genet. 2018; 19 :148–159. doi: 10.1038/nrg.2017.104. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Plomin R, von Stumm S. Polygenic scores: prediction versus explanation. Mol Psychiatry. 2022; 27 :49–52. doi: 10.1038/s41380-021-01348-y. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Polderman TJC, Benyamin B, de Leeuw CA, et al. Meta-analysis of the heritability of human traits based on fifty years of twin studies. Nat Genet. 2015; 47 :702–709. doi: 10.1038/ng.3285. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Purcell SM, Wray NR, Stone JL, et al. Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature. 2009; 460 :748–752. doi: 10.1038/nature08185. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Richmond RC, Davey Smith G. Mendelian randomization: concepts and scope. Cold Spring Harb Perspect Med. 2022; 12 :a040501. doi: 10.1101/cshperspect.a040501. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Rietveld CA, Medland SE, Derringer J, et al. GWAS of 126,559 individuals identifies genetic Variants Associated with Educational Attainment. Science. 2013; 340 :1467–1471. doi: 10.1126/science.1235488. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Risch N, Merikangas K. The future of genetic studies of complex human diseases. Science. 1996; 273 :1516–1517. doi: 10.1126/science.273.5281.1516. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ritchie S. Science fictions: exposing fraud, bias, negligence and hype in science. London: Vintage; 2021. [ Google Scholar ]
• Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA. 1977; 74 :5463–5467. doi: 10.1073/pnas.74.12.5463. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Savage JE, Jansen PR, Stringer S, et al. Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links to intelligence. Nat Genet. 2018; 50 :912–919. doi: 10.1038/s41588-018-0152-6. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Scarr S, Carter-Saltzman L. Genetics and intelligence. In: Sternberg RJ, editor. Handbook of human intelligence. Cambridge: Cambridge University Press; 1982. pp. 792–895. [ Google Scholar ]
• Scarr S, Weinberg RA. The influence of “family background” on intellectual attainment. Am Sociol Rev. 1978; 43 :674. doi: 10.2307/2094543. [ CrossRef ] [ Google Scholar ]
• Schoenfeldt LF. The hereditary components of the Project TALENT two-day test battery. Meas Evaluation Guidance. 1968; 1 :130–140. doi: 10.1080/00256307.1968.12022379. [ CrossRef ] [ Google Scholar ]
• Skodak M, Skeels HM. A final follow-up study of one hundred adopted children. The Pedagogical Seminary and Journal of Genetic Psychology. 1949; 75 :85–125. doi: 10.1080/08856559.1949.10533511. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Snyderman M, Rothman S. The IQ controversy, the media and public policy. New Brunswick, NJ: Transaction; 1990. [ Google Scholar ]
• Tallman GG. A comparative study of identical and non-identical twins with respect to intelligence resemblances. Teachers Coll Record. 1928; 29 :83–86. doi: 10.1177/016146812802900912. [ CrossRef ] [ Google Scholar ]
• The Wellcome Trust Case Control Consortium Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature. 2007; 447 :661–678. doi: 10.1038/nature05911. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Thorndike EL. The resemblance of siblings in intelligence. Teachers Coll Record. 1928; 29 :41–53. doi: 10.1177/016146812802900904. [ CrossRef ] [ Google Scholar ]
• Trzaskowski M, Davis OSP, DeFries JC, et al. DNA evidence for strong genome-wide pleiotropy of cognitive and learning abilities. Behav Genet. 2013; 43 :267–273. doi: 10.1007/s10519-013-9594-x. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Turkheimer E. Three laws of behavior genetics and what they mean. Curr Dir Psychol Sci. 2000; 9 :160–164. doi: 10.1111/1467-8721.00084. [ CrossRef ] [ Google Scholar ]
• Urbach P. Progress and degeneration in the “IQ debate” I and II. Br J Philosophical Sci. 1974; 25 :99–135. doi: 10.1093/bjps/25.2.99. [ CrossRef ] [ Google Scholar ]
• Visscher PM, Wray NR, Zhang Q, et al. 10 years of GWAS discovery: biology, function, and translation. Am J Hum Genet. 2017; 101 :5–22. doi: 10.1016/j.ajhg.2017.06.005. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Visscher PM, Yengo L, Cox NJ, Wray NR. Discovery and implications of polygenicity of common diseases. Science. 2021; 373 :1468–1473. doi: 10.1126/science.abi8206. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Wainschtein P, Jain D, Zheng Z, et al. Assessing the contribution of rare variants to complex trait heritability from whole-genome sequence data. Nat Genet. 2022; 54 :263–273. doi: 10.1038/s41588-021-00997-7. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Watson JB. Behaviorism, revised. Chicago: University of Chicago Press; 1930. [ Google Scholar ]
• Watson JD, Crick FHC. Genetical implications of the structure of deoxyribonucleic acid. Nature. 1953; 171 :964–967. doi: 10.1038/171964b0. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Wilson RS. The Louisville Twin Study: developmental synchronies in behavior. Child Dev. 1983; 54 :298. doi: 10.2307/1129693. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Woo SLC, Lidsky AS, Güttler F, et al. Cloned human phenylalanine hydroxylase gene allows prenatal diagnosis and carrier detection of classical phenylketonuria. Nature. 1983; 306 :151–155. doi: 10.1038/306151a0. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Woodworth RS. Heredity and environment: a criticla survey of recently published material on twins and foster children. New York: Social Science Research Council; 1941. [ Google Scholar ]
• Yang J, Manolio TA, Pasquale LR, et al. Genome partitioning of genetic variation for complex traits using common SNPs. Nat Genet. 2011; 43 :519–525. doi: 10.1038/ng.823. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Yengo L, Robinson MR, Keller MC, et al. Imprint of assortative mating on the human genome. Nat Hum Behav. 2018; 2 :948–954. doi: 10.1038/s41562-018-0476-3. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 27 March 2024

The effects of genetic and modifiable risk factors on brain regions vulnerable to ageing and disease

• Jordi Manuello   ORCID: orcid.org/0000-0002-9928-0924 1 , 2 ,
• Joosung Min   ORCID: orcid.org/0000-0002-5541-5014 3 ,
• Paul McCarthy 1 ,
• Fidel Alfaro-Almagro 1 ,
• Soojin Lee 1 , 4 ,
• Stephen Smith 1 ,
• Lloyd T. Elliott 3   na1 ,
• Anderson M. Winkler 5 , 6   na1 &
• Gwenaëlle Douaud   ORCID: orcid.org/0000-0003-1981-391X 1  

Nature Communications volume  15 , Article number:  2576 ( 2024 ) Cite this article

11k Accesses

2113 Altmetric

Metrics details

• Genetics research
• Neuroscience
• Risk factors

We have previously identified a network of higher-order brain regions particularly vulnerable to the ageing process, schizophrenia and Alzheimer’s disease. However, it remains unknown what the genetic influences on this fragile brain network are, and whether it can be altered by the most common modifiable risk factors for dementia. Here, in ~40,000 UK Biobank participants, we first show significant genome-wide associations between this brain network and seven genetic clusters implicated in cardiovascular deaths, schizophrenia, Alzheimer’s and Parkinson’s disease, and with the two antigens of the XG blood group located in the pseudoautosomal region of the sex chromosomes. We further reveal that the most deleterious modifiable risk factors for this vulnerable brain network are diabetes, nitrogen dioxide – a proxy for traffic-related air pollution – and alcohol intake frequency. The extent of these associations was uncovered by examining these modifiable risk factors in a single model to assess the unique contribution of each on the vulnerable brain network, above and beyond the dominating effects of age and sex. These results provide a comprehensive picture of the role played by genetic and modifiable risk factors on these fragile parts of the brain.

Similar content being viewed by others

Single-cell multiplex chromatin and RNA interactions in ageing human brain

Xingzhao Wen, Zhifei Luo, … Sheng Zhong

A concerted neuron–astrocyte program declines in ageing and schizophrenia

Emi Ling, James Nemesh, … Steven A. McCarroll

Cell subtype-specific effects of genetic variation in the Alzheimer’s disease brain

Masashi Fujita, Zongmei Gao, … Philip L. De Jager

Introduction

The development of preventative strategies based on modifying risk factors might prove to be a successful approach in ensuring healthy ageing. Factors particularly scrutinised in dementia and unhealthy ageing have included cerebrovascular factors such as high blood pressure, diabetes and obesity, but also lifestyle ones such as alcohol consumption, and protective factors such as exercise 1 . Assessing these modifiable risk factors together makes it possible to identify the unique contribution of each of these factors on the brain or on cognitive decline. A Lancet commission, updated in 2020 to include, e.g., pollution for its possible role in the incidence of dementia 2 , examined the relative impact of 12 modifiable risk factors for dementia, and showed that these 12 factors may account for 40% of the cases worldwide 3 . Conversely, genetic factors are non-modifiable in nature, but can inform us about the mechanisms underlying the phenotypes of interest. These mechanisms sometimes can be shared across these phenotypes. For instance, genetic overlap has been found for Alzheimer’s and Parkinson’s diseases at a locus in the MAPT region 4 . Likewise, one of the most pleiotropic variants, in the SLC39A8 / ZIP8 gene, shows genome-wide associations with both schizophrenia and fluid intelligence, amongst many other phenotypes 5 , 6 .

One way to objectively and robustly assess susceptibility for unhealthy ageing is to look non-invasively at brain imaging markers 7 . Using a data-driven approach on a lifespan cohort, we previously identified an ensemble of higher-order, ‘transmodal’ brain regions that degenerates earlier and faster than the rest of the brain 8 . The very same areas also develop relatively late during adolescence, thus supporting the ‘last in, first out’ (LIFO) hypothesis, which posits that the process of age-related brain decline mirrors developmental maturation. Importantly, this network of brain regions further demonstrated heightened vulnerability to schizophrenia and Alzheimer’s disease, two disorders that impact on brain structure during adolescence and ageing respectively. Accordingly, this LIFO network was strongly associated with cognitive traits whose impairment is specifically related to these two disorders, namely fluid intelligence and long-term memory 8 .

Here, our main objective was to assess both the genetic and modifiable risk factors’ contributions to the vulnerability of these most fragile parts of the brain. We conducted a genome-wide association study on a prospective cohort of nearly 40,000 participants of the UK Biobank study who had received brain imaging, and in total evaluated the association between the LIFO brain network and 161 modifiable risk factors, classified according to 15 broad categories: blood pressure, cholesterol, diabetes, weight, alcohol consumption, smoking, depressive mood, inflammation, pollution, hearing, sleep, socialisation, diet, physical activity and education.

The vulnerable LIFO brain network in UK Biobank

Similar to our previously observed results 8 , the loadings of the LIFO brain network, i.e., the normalised grey matter volume in the network after regressing out the effects of all the other brain maps (see Methods), demonstrated a strong quadratic association with age in the UK Biobank cohort of 39,676 participants ( R 2  = 0.30, P  < 2.23 × 10 −308 , Fig.  1 ). These higher-order regions thus show an accelerated decrease of grey matter volume compared with the rest of the brain. Furthermore, these areas define a network mainly involved in behavioural tasks related to execution, working memory, and attention (Fig.  1 , Supplementary Information ).

Top left, spatial map of the LIFO network (in red-yellow, thresholded at Z  > 4 for visualisation) used to extract the loadings from every scanned participant from UK Biobank ( n  = 39,676). Top right, these LIFO loadings (in arbitrary units) show a strong quadratic association with age in the UK Biobank cohort, i.e. grey matter volume decreases quadratically with older age in these specific regions ( R 2  = 0.30, P  < 2.23 × 10 −308 ; inset: residual scatterplot). Bottom, the vulnerable network appears to encompass areas mainly involved in execution, working memory, and attention (using the BrainMap taxonomy 60 , and with the LIFO brain network thresholded at both Z  = 4 and Z  = 10, see  Supplementary Information ).

Genetic influences over the vulnerable LIFO brain network

Using a minor allele frequency filter of 1% and a –log 10 (P) threshold of 7.5, we found, in the 39,676 participants, genome-wide associations between the LIFO brain network and seven genetic clusters whose top variants were all replicated (Table  1 /Supplementary Data  1 , Fig.  2 ).

Top row, Manhattan plot showing the 7 significant genetic clusters associated with the LIFO brain network (–log 10 ( P ) > 7.5). Second and third rows, regional association plots of the top variants for each of the 5 autosomal genetic clusters: rs6540873 on chromosome (Chr) 1 ( KCNK2 ), rs13107325 on Chr4 ( SLC39A8 ), rs2677109 on Chr6 ( RUNX2 ) (as a proxy in high LD R 2  = 0.86 with indel 6:45442860_TA_T), rs12146713 on Chr12 ( NUAK1 ), and rs2532395 on Chr17 ( MAPT , KANSL1 )(highest variant after tri-allelic rs2693333; see Supplementary Data  4 for a complete list of significant variants in this 5th MAPT genetic cluster). Bottom row, regional association plots of the top variants for the two genetic clusters in the pseudo-autosomal region PAR1 of the X chromosome: rs312238 ( XG , CD99 ) and rs2857316 ( XG )(UK Biobank has no genotyped variants on the 3’ side). Based on Human Genome build hg19. P -values are derived from a two-sided linear association test.

The first autosomal genetic cluster, on chromosome 1, included two variants (lead variant: rs6540873, β  = 0.06, P  = 1.71 × 10 −8 , and rs1452628, with posterior probabilities of inclusion in the causal variant set of 0.56 and 0.45, respectively) close to, and eQTL of, KCNK2 ( TREK1 ). This gene regulates immune-cell trafficking into the central nervous system, controls inflammation, and plays a major role in the neuroprotection against ischemia. Of relevance, these two loci are in particular related in UK Biobank participants with the amount of alcohol consumed, insulin levels, inflammation with interleukin-8 levels, as well as, crucially, with late-onset Alzheimer’s disease (Table  1 /Supplementary Data  1 ).

The second autosomal genetic cluster on chromosome 4 was made of 7 loci, with the lead variant rs13107325 in an exon of SLC39A8/ZIP8 ( β  = 0.14, P  = 2.82 × 10 −13 , posterior probability: 0.99). This locus is one of the most pleiotropic SNPs identified in GWAS, and is, amongst many other associations, related in UK Biobank with cholesterol, blood pressure, weight, inflammation with C-reactive proteins levels, diabetes with insuline-like growth factor 1 levels, alcohol intake, sleep duration, and cognitive performance/impairment, including prospective memory (Table 1 /Supplementary Data  1 ).

The third locus was an indel in chromosome 6 in an intron, and eQTL, of RUNX2 (rs35187443, β  = 0.06, P  = 9.03 × 10 −9 ), which plays a key role in differentiating osteoblasts, and has been very recently shown to limit neurogenesis and oligodendrogenesis in a cellular model of Alzheimer’s disease 9 .

The fourth locus was a SNP in chromosome 12, in an intron of NUAK1 (rs12146713, β  = −0.10, P  = 1.26 × 10 −9 ), and remarkably its top association in UK Biobank was with the contrast between schizophrenia and major depressive disorder 10 , and it was also associated with insulin-like growth factor 1 levels (Table 1 /Supplementary Data  1 ).

The final genetic autosomal genetic cluster was made of 3,906 variants in the MAPT region. Its lead non-triallelic variant, rs2532395 ( β  = −0.09, P  = 3.56 × 10 −15 ) was more specifically <10 kb from KANSL1 and an eQTL of KANSL1 , MAPT and other genes in brain tissues (Table 1 /Supplementary Data  1 , Supplementary Data 4 ). This locus was also associated in UK Biobank with tiredness and alcohol intake. MAPT is in 17q21.31, a chromosomal band involved with a common chromosome 17 inversion 11 . Adding chromosome 17 inversion status as a confounder reduced the significance of the association ( β  = −0.15, P  = 8.45 × 10 −3 ). Since the genotype for rs2532395 was also strongly correlated with chromosome 17 inversion in our dataset (Pearson correlation r  = 0.98, P  < 2 × 10 −16 ), this would suggest that the association between MAPT and the LIFO network is not independent from chromosome 17 inversion. As this extended genetic region is known for its pathological association with many neurodegenerative disorders including Alzheimer’s disease, we investigated whether the LIFO brain regions mediated the effect of the MAPT genetic cluster (using the lead bi-allelic variant rs2532395) on Alzheimer’s disease (see Methods). Despite small average causal mediated effect (ACME) sizes, we found a significant effect for both the dominant model (ACME β  = 1.16 × 10 −4 ; 95% CI = [5.19 × 10 −5 , 1.99 × 10 −4 ]; P  = 4 × 10 −5 ) and the recessive model (ACME β  = 1.55 × 10 −4 ; 95% CI = [3.96 × 10 −5 , 3.74 × 10 −4 ]; P  = 4 × 10 −5 ; full output of the mediation package on the dominant and recessive models in  Supplementary Information ).

The two last genetic clusters of 8 and 9 variants respectively were found on the X chromosome, notably in a pseudo-autosomal region (PAR1), which is interestingly hit at a higher rate than the rest of the genome ( P  = 1.56 × 10 −5 , see  Supplementary Information ). The top variants for these clusters were related to two homologous genes coding for the two antigens of the XG blood group: rs312238 ( β  = −0.05, P  = 1.77 × 10 −10 ) ~ 10 kb from, and an eQTL of, CD99/MIC2 , and rs2857316 ( β  = −0.08, P  = 2.27 × 10 −29 ) in an intron and eQTL of XG  (Table 1 /Supplementary Data  1 ). Since chromosome X has hardly been explored, we carried out our own association analyses between these two top variants and non-imaging variables in UK Biobank. Intriguingly, the first of these two PAR1 loci, rs312238, was found to be significantly associated in the genotyped participants who had not been scanned (out-of-sample analysis in n  = 374,230 UK Biobank participants) with nitrogen dioxide air pollution, our ‘best’ MRF for pollution (see below), and many other environmental, socioeconomic, and early life factors (such as urban or rural setting, distance from the coast, place of birth, number of siblings, breastfed as a baby, maternal smoking around birth), as well as health outcomes (Supplementary Data  2 ). In particular, amongst the more easily interpretable findings of the most associated variables with rs312238, the T allele of this locus was associated with two increased measures of deprivation and/or disability (worse socioeconomic status), the ‘Townsend deprivation index’ and the ‘Health score’, but also with ‘Nitrogen dioxide air pollution’, ‘Maternal smoking around birth’, as well as ‘Number of full brothers’ and ‘Number of full sisters’, thus showing consistent signs of association between this variant and these phenotypes.

We found that the heritability of the LIFO network was significant, with h 2  = 0.15 (se = 0.01). The genetic co-heritability between the LIFO network and Alzheimer’s disease or schizophrenia was not statistically significant (coefficient of co-heritability = −0.12, se = 0.10; P  = 0.23; coefficient of co-heritability = −0.16, se = 0.04, P  = 0.07, respectively).

Modifiable risk factors’ associations with the vulnerable LIFO brain network

Including the modifiable risk factors (MRFs) in a single general linear model allows us to assess the unique contribution of each factor on the LIFO brain network. Not all UK Biobank participants have data available for all of the MRF variables however. An analysis limited to those with complete data for all MRFs would be biased, and based on a relatively small, low-powered sample. We addressed this issue via a two-stage analysis in which: (i) we first identified which variable within each of the 15 MRF categories best represented associations of that category with the LIFO brain network loadings (based on two criteria: significance and <5% missing values), (ii) we investigated the unique contribution of that MRF category, over and above all other categories and the dominating effects of age and sex, to the LIFO loadings.

From the first stage of our analysis, 12 of the 15 categories of MRFs had at least one ‘best’ MRF, i.e., with a significant effect on the LIFO brain network and enough non-missing values across all scanned participants to be investigated further (Table  2 /Supplementary Data  3 ). The contribution of the MRFs on the vulnerable brain network differed vastly depending on whether confounding effects of age, sex and head size were taken into account. The effect size and significance of some MRFs diminished because of some clear collinearity with the confounders. For instance, for the category of blood pressure, the most significant MRF was first “systolic blood pressure, automatic (second) reading” ( r  = −0.20, P  < 2.23 × 10 −308 ), but after regressing out the confounders, the ‘best’ MRF for this category was “medication for blood pressure” ( r  = −0.05, P  = 7.55 × 10 −22 ). Conversely, regressing out the effects of age served to unmask the significant deleterious effects of pollution on the vulnerable brain regions, such as nitrogen dioxide air pollution or particulate matter air pollution (Table  2 /Supplementary Data  3 ).

When considered together in a single model in the second stage of the analysis, 3 best MRFs had an effect on the LIFO brain network that remained significant beyond the dominating effects of age and sex, and of the 9 other best MRFs: diabetes (“diabetes diagnosed by doctor”, r  = −0.05, P  = 1.13 × 10 −24 ), pollution (“nitrogen dioxide air pollution in 2005”, r  = −0.05, P  = 5.39 × 10 −20 ) and alcohol (“alcohol intake frequency”, r  = −0.04, P  = 3.81 × 10 −17 ) (Table  3 ). No MRFs showed any bias in their sub-sampling distribution, i.e., any significant difference between the original sample and the reduced sample of 35,527 participants who had values for all 18 variables considered (the 12 best MRFs and 6 confounders: age, sex, age 2 , age × sex, age 2  × sex, head size; Supplementary Information ). In total, the 12 best MRFs explained 1.5% of the effect on the vulnerable brain network ( F 12;35509  = 43.5).

While 6 out of the 7 genetic clusters associated with the LIFO network were correlated with many variables related to each of the 15 MRF categories, including diabetes, alcohol consumption and traffic pollution (Supplementary Data  1 ), we also found some genetic overlap between the very specific best MRF of “alcohol intake frequency” and the LIFO network in the pleiotropic rs13107325 variant (cluster 2), as well as rs17690703, part of the large genetic cluster 5 in MAPT (Supplementary Data  4 ). No genetic overlap was found for the precise “nitrogen dioxide air pollution in 2005” or “diabetes diagnosed by doctor”, nor for approximate variables.

This study reveals, in a cohort of nearly 40,000 UK Biobank participants, the genetic and modifiable risk factors’ associations with brain regions in a ‘last in, first out’ (LIFO) network that show earlier and accelerated ageing and are particularly vulnerable to disease processes such as that of Alzheimer’s disease 8 . Seven genetic clusters, two of which in the pseudo-autosomal region of the sex chromosomes coding for two antigens of the XG blood system, were found significantly associated and replicated genome-wide. In addition, after accounting for age and sex effects, diabetes, traffic-related pollution and alcohol were the most deleterious modifiable risk factors (MRFs) on these particularly vulnerable brain regions.

Three lead variants for our significant genetic clusters have been previously associated with ageing-related brain imaging measures in recent studies: one, in cluster 1, an eQTL of KCNK2 ( TREK1 ) 12 , 13 , whose increase in expression mediates neuroprotection during ischemia 14 , the ubiquitous rs13107325 (cluster 2), and one, in cluster 4, in an intron of NUAK1 ( ARK5 ) 15 , 16 , 17 , which has been associated with tau pathology 18 (Table  1 /Supplementary Data  1 ). On the other hand, of the seven genetic clusters, three were entirely novel (clusters 3, 6 and 7), and not found in other brain imaging studies, including our most recent work that expanded on our previous GWAS of all of the brain IDPs available in UK Biobank 19 by including more participants—in fact, the same number of participants as analysed in this present work—and, crucially, by also including the X chromosome 20 (Table  1 /Supplementary Data  1 ). This suggests that, beyond the genetic hits that were meaningfully associated with the LIFO brain network and an array of relevant risk factors, lifestyle variables and brain disorders, and found in a few other imaging GWAS, some of the genetic underpinnings of the LIFO network are intrinsically specific to it and to no other pre-existing imaging phenotype.

All five autosomal genetic clusters identified through the GWAS of the LIFO phenotype had relevant associations with risk factors for dementia (Results; Supplementary Data  1 ), including precisely two of the best MRFs (for clusters 2 and 5), and three of them directly related in UK Biobank to the two diseases showing a pattern of brain abnormalities following the LIFO network: schizophrenia (clusters 2 and 4) and Alzheimer’s disease (cluster 1) (Supplementary Data  1 ). In particular, cluster 2 has its lead variant rs13107325 in an exon of one of the most pleiotropic genes ZIP8 , which codes for a zinc and metal transporter. Considering the vulnerability of the LIFO brain network to adolescent-onset schizophrenia and its significant association with fluid intelligence that we previously demonstrated 8 , it is notable that this variant has been associated genome-wide with schizophrenia 6 , as well as intelligence, educational attainment and mathematics ability 5 , 21 . In line with the LIFO brain network being both prone to accelerated ageing and susceptible to Alzheimer’s disease, this genetic locus has also been associated genome-wide with well-known risk factors for dementia. These comprise alcohol—including the exact same variable of “alcohol intake frequency” as identified as one of the best MRFs—cholesterol, weight, sleep—including “sleep duration”—and blood pressure 22 , 23 , 24 , 25 , 26 , all of which significantly contribute to modulating the LIFO brain network when considered separately (Table  2 /Supplementary Data  3 ). Of relevance, this genetic locus is also associated to an increased risk of cardiovascular death 27 . Cluster 5, a large genetic cluster in the MAPT region (Microtubule-Associated Protein Tau), comprised in total 3906 significant variants (Supplementary Data  4 ). This genetic region plays a role in various neurodegenerative disorders related to mutations of the protein tau, such as frontotemporal dementia 28 and progressive supranuclear palsy 29 , but also, of particular pertinence to the LIFO brain network, Alzheimer’s and Parkinson’s disease, with a genetic overlap between these two diseases in a locus included in our significant cluster 5 (rs393152, β  = −0.09, P  = 6.35 × 10 −14 ) 4 . Despite the relatively low number of people with diagnosed Alzheimer’s disease in the genetic discovery cohort, we were able to establish—albeit with small effect sizes—a significant mediation role for the LIFO brain regions between the lead bi-allelic variant for cluster 5 and this Alzheimer’s diagnosis, suggesting once more the importance played by these vulnerable brain areas in unhealthy ageing.

Finally, of the seven clusters, two were located in the pseudo-autosomal region (PAR1) of the sex chromosomes corresponding to the genes XG and CD99 , coding for the two antigens of the XG blood group. This blood group system has been largely neglected, its main contribution related to the mapping of the X chromosome itself, and its clinical role remains elusive 30 . In order to investigate further the possible role of these two variants of the XG blood group, we examined out-of-sample their associations with thousands of non-imaging phenotypes. This analysis revealed that the first of these two loci was significantly and consistently associated with early life factors, environmental factors and health outcomes, including particulate matter and nitrogen dioxide air pollution, the second most deleterious MRF to the LIFO brain network (Supplementary Data  2 ). Whether these associations are due to stratification or genotyping artefacts, or to the fact that this specific variant, which is inherited from a parent, has a parental impact that modulates the effect of early life environment of the UK Biobank participants, the so-called “nature of nurture”, will need further investigation 31 .

Intriguingly, an analysis revealed that the genes involved in the loci associated with the LIFO network (Table  1 /Supplementary Data  1 ) are enriched for the gene ontology terms of leucocyte extravasation, namely “positive regulation of neutrophil extravasation” ( P  = 4.75 × 10 −6 ) and “T cell extravasation” ( P  = 4.75 × 10 −6 ). This result held when removing the genes included in the MAPT extended region (with P  = 2.54 × 10 −6 and P  = 2.54 × 10 −6 , respectively). Leucocyte extravasation facilitates the immune and inflammatory response, and there has been renewed focus on the fact that a breakdown of the blood-brain barrier together with leukocyte extravasation might contribute to both Alzheimer’s disease and schizophrenia 32 , 33 . In line with the enrichment findings, 4 out of the 7 genetic clusters associated with the LIFO network are correlated in UK Biobank blood assays with percentage or count of immune cells (neutrophil, lymphocyte, platelet, monocyte, etc.; Supplementary Data  1 ).

Regarding MRFs’ effects on the LIFO brain network, diabetes and alcohol consumption have been consistently shown to be associated with both cerebral and cognitive decline 34 , 35 . On the other hand, pollution—and notably that of nitrogen oxides—has emerged more recently as a potential MRF for dementia 2 , 36 . In particular, the increase of dementia risk due to nitrogen oxide pollution, a proxy for traffic-related air pollution, seems to be enhanced by cardiovascular disease 37 . In this study, we found that nitrogen dioxide pollution has one of the most deleterious effects onto the fragile LIFO brain regions. This effect could only be unmasked by regressing out the effects of age and sex, as traffic-related air pollution is modestly inversely-correlated with age (Supplementary Data  5 ). It is also worth noting that including age and sex as confounding variables in the first stage of our analysis reduced considerably the contribution of what had appeared at first—before regression—as the most harmful risk factors: blood pressure, cholesterol and weight (Table  2 /Supplementary Data  3 ). Furthermore, the benefit of examining these MRFs in a single model in the second stage of our analysis is that we can assess the unique contribution of each of these factors on the LIFO brain network; in doing so, blood pressure, cholesterol and weight were no longer significant (Table  3 ).

One defining characteristic of the LIFO brain network is how much age explains its variance. Indeed, in the dataset covering most of the lifespan that was initially used to identify the LIFO and spatially define it 8 , age explained 50%. In the UK Biobank imaging project, where imaged participants are over 45 years old, age explained 30% (Fig.  1 ). It is thus perhaps unsurprising that, while the explained variance by each of the MRFs varies widely (Table  2 /Supplementary Data  3 ), it reduces notably once the effect of age and other confounders has been regressed out (without confounders included in the model: maximum 8.4%; with confounders: maximum 0.5%). Combined, the 12 best MRFs explained a significant 1.5% of the effect on the vulnerable brain network after regressing out age, head size and sex effects. Regarding the genetic hits, we found a significant heritability with h 2  = 0.15, in keeping with our results for structural brain phenotypes (except for subcortical and global brain volumes, which demonstrate higher heritability 19 ).

The uniqueness of this study relies on the fact that we combined the strengths of two different cohorts: the first, which revealed the LIFO grey matter network, is lifespan, demonstrating the mirroring of developmental and ageing processes in the LIFO brain areas, something that could never be achieved with UK Biobank because of its limited age range. Of note, for this initial work with the lifespan cohort 8 , we not only included grey matter partial volume images, as done in this current study, but also Freesurfer information of cortical thickness and surface area. The LIFO network showed no contribution from Freesurfer cortical thickness or area. This might hint at processes that only partial volume maps are able to detect due to the LIFO network’s specific localisation, including in the cerebellum and subcortical structures, which are not included in the area and thickness surface methods from Freesurfer.

Limitations of our study pertain to the nature of the data itself and the way each variable is encoded in the UK Biobank (binary, ordinal, categorical, continuous), the number of missing values, what is offered as variables for each modifiable risk factor category (e.g. we chose not to create any compound variables, such as the ratio of cholesterol levels or systolic and diastolic blood pressures), and the curation of each of these variables. Some of the factors might be proxies for another category, but including the ‘best’ ones in a single model alleviate these issues to some extent. Another limitation is the assumption in our models that each risk factor has a linear, additive effect on the vulnerable LIFO brain network. It is also important to note that cross-sectional and longitudinal patterns of brain ageing can differ, as has been shown for instance for adult span trajectories of episodic and semantic memory, especially in younger adults 38 . A recent study has also demonstrated a specific ‘brain age’ imaging measure to be more related to early life influences on brain structure than within-person rates of change in the ageing brain 39 . Further work will be needed to establish how the LIFO network data changes in terms of within-person trends, for instance by investigating the growing UK Biobank longitudinal imaging database. While we took care of assessing the replicability of our genetic results by randomly assigning a third of our dataset for such purposes (all our significant genetic hits were replicated), this was performed within the UK Biobank cohort that exhibits well-documented biases, being well-educated, less deprived, and healthier than the general population, especially for its imaging arm 40 . Independent replications will be needed to confirm the existence of the LIFO-associated genetic loci.

In conclusion, our study reveals the modifiable and non-modifiable factors associated with some of the most fragile parts of the brain particularly vulnerable to ageing and disease process. It shows that, above and beyond the effect of age and sex, the most deleterious modifiable risk factors to this brain network of higher-order regions are diabetes, pollution and alcohol intake. Genetic factors are related to immune and inflammatory response, tau pathology, metal transport and vascular dysfunction, as well as to the XG blood group system from the pseudo-autosomal region of the sex chromosomes, and meaningfully associated with relevant modifiable risk factors for dementia. The unprecedented genome-wide discovery of the two variants on the sex chromosomes in this relatively unexplored blood group opens the way for further investigation into its possible role in underlying unhealthy ageing.

Supplementary Information is available for this paper.

For the present work the imaging cohort of UK Biobank was used and we included 39,676 subjects who had been scanned and for whom the brain scans had been preprocessed at the time of the final set of analyses (M/F 47–53%; 44–82 years, mean age 64 ± 7 years; as of October 2020) 41 , 42 . Structural T1-weighted scans for each participant were processed using the FSL-VBM automated tool to extract their grey matter map 43 , 44 . The ‘last in, first out’ (LIFO) network of mainly higher-order brain regions was initially identified by performing a linked independent component analysis on the grey matter images of another, lifespan observational cohort of 484 subjects 8 , 45 , 46 . This map of interest, along with the other 69 generated by the analysis, was first realigned to the UK Biobank ‘standard’ space defined by the grey matter average across the first 15,000 participants, then regressed into the UK Biobank participants’ grey matter data, to extract weighted average values of grey matter normalised volume inside each of the z-maps, using the z-score as weighting factor. This made it possible to assess the unique contribution of this specific LIFO map, above and beyond all the rest of the brain represented in the other 69 maps. At the end of this process, we obtained a single imaging measure for each of the 39,676 participants, i.e. a ‘loading’ corresponding to their amount of grey matter normalised volume in the LIFO brain network.

Human participants: UK Biobank has approval from the North West Multi-Centre Research Ethics Committee (MREC) to obtain and disseminate data and samples from the participants ( http://www.ukbiobank.ac.uk/ethics/ ), and these ethical regulations cover the work in this study. Written informed consent was obtained from all of the participants.

Modifiable risk factors selection

The following 15 categories of modifiable risk factors (MRFs) for dementia were investigated based on previous literature: blood pressure, diabetes, cholesterol, weight, alcohol, smoking, depression, hearing, inflammation, pollution, sleep, exercise, diet/supplementation, socialisation, and education. These included well-documented cerebrovascular risk factors, and in particular included all of the 12 modifiable risk factors considered in the updated Lancet commission on dementia, with the sole exception of traumatic brain injury 3 . For each category, several MRF variables from UK Biobank were very minimally pre-processed ( Supplementary Information ). In total, 161 MRF variables were obtained. To optimise the interpretability of the results, and to be able to relate them to previous findings, we did not carry out any data reduction, which would have prevented us from identifying exactly which variable—and subsequently, which genetic component for this specific variable—contribute to the effect. For these same reasons, we did not create any compound variable.

Statistical analyses

Genome-wide association study.

We followed the same protocol we had developed for the first genome-wide association study (GWAS) with imaging carried out on UK Biobank 19 . Briefly, we examined imputed UK Biobank genotype data 47 , and restricted the analysis to samples that were unrelated (thereby setting aside only ~450 participants), without aneuploidy and with recent UK ancestry. To account for population stratification, 40 genetic principal components were used in the genetic association tests as is recommended for UK Biobank genetic studies 19 , 20 , 47 . We excluded genetic variants with minor allele frequency <0.01 or INFO score <0.03 or Hardy-Weinberg equilibrium –log 10 ( P ) > 7. We then randomly split the samples into a discovery set with 2/3 of the samples ( n  = 22,128) and a replication set with 1/3 of the samples ( n  = 11,083). We also examined the X chromosome with the same filters, additionally excluding participants with sex chromosome aneuploidy: 12 in non-pseudoautosomal region (PAR) and 9 in PAR for the discovery set, 3 in non-PAR and 6 in PAR for the replication set. Variants were considered significant at –log 10 ( P ) > 7.5, and replicated at P  < 0.05.

Modifiable risk factor study

In the first stage, the general linear model was used to investigate, separately, the association between each of these 161 MRFs and the LIFO network loadings in all the scanned UK Biobank participants ( n  = 39,676). We ran each model twice: once as is, and once adding 6 confounders: age, age 2 , sex, age × sex, age 2 × sex, and head size, to estimate the contribution of these MRFs on the LIFO network above and beyond the dominating effects of age and sex. Sex was based on the population characteristics entry of UK Biobank. This is a mixture of the sex the NHS had recorded for the participant at recruitment, and updated self-reported sex. For the GWAS, both sex and genetic sex were used (the sample was excluded in case of a mismatch). In total, 32 variables tailored to structural imaging had been considered as possible confounders, and we retained those with the strongest association ( R 2  ≥ 0.01; see  Supplementary Information ). Socioeconomic status via the Townsend deprivation index was also considered as a possible confounding variable but explained little variance ( R 2  < 0.001) and thus was not included as a confounder.

MRFs were not considered further if they were not significant—not surviving Bonferroni-correction, i.e., P  > 1.55 × 10 −4 —and if more than 5% of the subjects had their MRF values missing. For each category, a single ‘best’ MRF was then selected as the variable with the highest R 2 among those remaining, after regressing out the confounding effects of age and sex.

In the second stage, all these best MRFs were then included in a single general linear model, together with the same 6 confounders used in the first stage, to assess the unique contribution of each factor on the LIFO brain network loadings. A prerequisite to carry out this single general linear model analysis was to only include participants who would have values for all best MRFs and confounders. This explains the additional criterion of only including MRFs that had no more than 5% of values missing, to ensure that the final sample of participants who had values for all these best and confounding factors would not be biased compared with the original sample—something we formally tested (see  Supplementary Information )—especially as data are not missing at random in UK Biobank, and exhibit some genetic structure 48 . The sample was therefore reduced to a total of 35,527 participants for this second stage analysis (M/F 17,290–18,237; 45–82 years, mean 64 ± 7 years). The effect of these best MRFs taken altogether was considered significant with a very conservative Bonferroni correction for multiple comparisons across all combinations of every possible MRF from each of the initial 15 MRF categories ( P  < 4.62 × 10 −17 , see  Supplementary Information for more details). In addition, both full and partial correlations were computed for the same set of best MRFs and confounders, in order to assess possible relationships between variables.

Post hoc genetic analyses

Chromosome 17 inversion.

We investigated chromosome 17 inversion status of the participants in the discovery cohort by considering their genotype on 32 variants that tag chromosome 17 inversion according to Steinberg et al. 11 . Of these 32 variants, 24 were present in our genetic data. We labelled the participants homozygous inverted, heterozygous, or homozygous direct (not inverted) when all 24 of these alleles indicated the same zygosity. This yielded an unambiguous inversion status for 21,969 participants (99% of the discovery cohort). To examine if the association between the non-triallelic lead variant of the MAPT genetic cluster (rs2532395, Table  1 /Supplementary Data  1 ) and the LIFO network was independent from this common inversion, we determined inversion/direct status of the discovery cohort and: 1. repeated the association test between rs2532395 and the LIFO phenotype, with chromosome 17 inversion status added as a confounder; and 2. correlated the genotype for rs2532395 with chromosome 17 inversion.

Causality within each genetic cluster

We used CAVIAR (Causal Variants Identification in Associated Regions 49 ) to assess causality of variants that passed the genome-wide significance threshold in each of the genetic clusters we report. CAVIAR uses a Bayesian model and the local linkage disequilibrium structure to assign posterior probabilities of causality to each variant in a region, given summary statistics for an association. We did not perform CAVIAR analysis on the genetic cluster on chromosome 17, as its non-triallelic lead variant (rs2532395) was strongly correlated with chromosome 17 inversion, and the LD matrix was large and low rank. We excluded the X chromosome loci from this analysis due to the difficulty in assessing LD in this chromosome.

Enrichment analysis

Based on the genes listed in the ‘Genes’ column of Table  1 /Supplementary Data  1 , we performed an enrichment analysis for the genes associated with the LIFO brain network using PANTHER 50 . PANTHER determines whether a gene function is overrepresented in a set of genes, according to the gene ontology consortium 51 , 52 .

Mediation analysis between MAPT top variant and Alzheimer’s disease, via the LIFO brain network

As the gene MAPT is associated with Alzheimer’s disease, and as we found a significant association between MAPT and the LIFO brain network, we examined to what extent the effect of MAPT is mediated by the LIFO brain regions. We conducted a mediation analysis using the counterfactual framework in which the average indirect effect of the treatment on the outcome through the mediator is nonparametrically identified (version 4.5.0 of the R package ‘mediation' 53 ). This is a general approach that encompasses the classical linear structural equation modelling framework for causal mediation, allowing both linear and non-linear relationships. In this analysis, the genotype for the lead bi-allelic variant of the MAPT association was used as the treatment, the LIFO loadings as the mediator, and Alzheimer’s disease diagnosis as the outcome.

From the ~43 K UK Biobank participants who had been scanned, we searched for those who had been diagnosed with Alzheimer’s disease specifically, regardless of whether this diagnosis occurred before, or after their brain scans. Based on hospital inpatient records (ICD10: F000, F001, F002, F009, G300, G301, G308, and G309 and ICD9: 3310) and primary care (GP) data (Eu00., Eu000, Eu001, Eu002, Eu00z, F110., F1100, F1101, Fyu30, X002x, X002y, X002z, X0030, X0031, X0032, X0033, XaIKB, XaIKC, and XE17j), we identified 65 such cases— UK Biobank being healthier than the general population, and those scanned showing an even stronger healthy bias—of which 34 were included in the discovery set after QC.

We considered two conditions for the effect of the treatment on the outcome. First, a dominant condition in which the minor allele is assumed to be dominant and for which at least one copy of the minor allele is considered treated. Second, a recessive condition in which the minor allele is assumed to be recessive. We considered that either condition was nominally significant if the confidence interval of the average causal mediated effect did not intersect zero, and had an associated P  < 0.05 ÷ 2 (correcting for the two conditions). We assessed confidence intervals and P -values using 50,000 bootstrapped samples.

Associations between the LIFO brain network’s genetic hits and the MRFs

First, we reported in Table  1 / Supplementary Data  1 the significant associations between the LIFO genetic hits and UK Biobank variables related to the 15 categories listed for the MRFs. For this, we used the Open Targets Genetics website, which reports the GWAS carried out in UK Biobank ( https://genetics .opentargets.org/ ). Second, we assessed whether there was any genetic overlap between the known genetic components of the 3 best MRFs and the LIFO phenotype. Again, we used the Open Targets Genetics website outputs for these 3 very specific UK Biobank variables, and compared the significant hits for these 3 best MRFs within ±250 kbp of, or in high LD (>0.8) with, our own LIFO variants. If reported hits were limited, we also searched online for GWAS done on similar variables. Finally, we also included the list of significant hits for diabetes 54 , which focused on a potential genetic overlap between diabetes and Alzheimer’s disease.

Post hoc association for the sex chromosomes variants

The allele counts of each participant for two specific significant variants of the sex chromosomes not—or hardly—available in open databases such as https://genetics.opentargets.org/ 55 were further associated out-of-sample with all non-imaging phenotypes of UK Biobank ( n  = 16,924). This analysis was carried out in the entire genotyped, quality-controlled sample where participants who had been scanned were removed (final sample: 374,230 participants), taking into account the population structure (40 genetic principal components), as well as the confounding effects of age, sex, age x sex, age 2 and age 2 x sex. Results were corrected for multiple comparisons across all non-imaging phenotypes and the two variants.

Heritability

We examined the heritability of the LIFO phenotype, and the coheritability between the LIFO network and Alzheimer’s disease or schizophrenia using LDSC 56 . This method uses regression on summary statistics to determine narrow sense heritability h 2 of a trait, or the shared genetic architecture between two traits. LDSC corrects for bias LD structure using LD calculated from a reference panel (we used LD from the Thousand Genomes Project Phase 1 57 ). We obtained summary statistics for a meta-analysis of Alzheimer’s disease involving 71,880 cases and 383,378 controls 58 . The number of genetic variants in the intersection between the summary statistics was 1,122,435. For schizophrenia, the summary statistics were obtained from a meta-analysis involving 53,386 cases and 77,258 controls 59 . A total of 1,171,319 genetic variants were in the intersection with the summary statistics for LIFO. For both Alzheimer’s and schizophrenia, the X chromosome was not included in the heritability calculation, as it was excluded from the meta-analysis that we sourced the summary statistics from.

Reproducibility

No data was excluded for the MRF analyses. For the genetic analyses, these were restricted to samples that were unrelated, without aneuploidy and with recent UK ancestry (see above).

No statistical method was used to predetermine sample size. The experiments were not randomised. The Investigators were not blinded to allocation during experiments and outcome assessment.

Reporting summary

Further information on research design is available in the  Nature Portfolio Reporting Summary linked to this article.

Data availability

All the FLICA decomposition maps − including the LIFO grey matter network − in UK Biobank standard space, the UK Biobank grey matter template, scripts, and the LIFO loadings for all of the participants are freely available on a dedicated webpage: open.win.ox.ac.uk/pages/douaud/ukb-lifo-flica/ .

Baumgart, M. et al. Summary of the evidence on modifiable risk factors for cognitive decline and dementia: a population-based perspective. Alzheimers Dement. 11 , 718–726 (2015).

Article   PubMed   Google Scholar  

Chen, H. et al. Exposure to ambient air pollution and the incidence of dementia: a population-based cohort study. Environ. Int. 108 , 271–277 (2017).

Article   CAS   PubMed   Google Scholar  

Livingston, G. et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet 396 , 413–446 (2020).

Article   PubMed   PubMed Central   Google Scholar  

Desikan, R. S. et al. Genetic overlap between Alzheimer’s disease and Parkinson’s disease at the MAPT locus. Mol. Psychiatry 20 , 1588–1595 (2015).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Hill, W. D. et al. A combined analysis of genetically correlated traits identifies 187 loci and a role for neurogenesis and myelination in intelligence. Mol. Psychiatry 24 , 169–181 (2019).

Pardinas, A. F. et al. Common schizophrenia alleles are enriched in mutation-intolerant genes and in regions under strong background selection. Nat. Genet. 50 , 381–389 (2018).

Douaud, G. et al. Preventing Alzheimer’s disease-related gray matter atrophy by B-vitamin treatment. Proc. Natl Acad. Sci. USA 110 , 9523–9528 (2013).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Douaud, G. et al. A common brain network links development, aging, and vulnerability to disease. Proc. Natl Acad. Sci. USA 111 , 17648–17653 (2014).

Nakatsu, D. et al. BMP4-SMAD1/5/9-RUNX2 pathway activation inhibits neurogenesis and oligodendrogenesis in Alzheimer’s patients’ iPSCs in senescence-related conditions. Stem Cell Rep. 18 , 1246 (2023).

Article   CAS   Google Scholar  

Peyrot, W. J. & Price, A. L. Identifying loci with different allele frequencies among cases of eight psychiatric disorders using CC-GWAS. Nat. Genet. 53 , 445–454 (2021).

Steinberg, K. M. et al. Structural diversity and African origin of the 17q21.31 inversion polymorphism. Nat. Genet. 44 , 872–880 (2012).

Le Guen, Y. et al. eQTL of KCNK2 regionally influences the brain sulcal widening: evidence from 15,597 UK Biobank participants with neuroimaging data. Brain Struct. Funct. 224 , 847–857 (2019).

Jonsson, B. A. et al. Brain age prediction using deep learning uncovers associated sequence variants. Nat. Commun. 10 , 5409 (2019).

Heurteaux, C. et al. TREK-1, a K+ channel involved in neuroprotection and general anesthesia. EMBO J. 23 , 2684–2695 (2004).

Vojinovic, D. et al. Genome-wide association study of 23,500 individuals identifies 7 loci associated with brain ventricular volume. Nat. Commun. 9 , 3945 (2018).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Zhao, B. et al. Genome-wide association analysis of 19,629 individuals identifies variants influencing regional brain volumes and refines their genetic co-architecture with cognitive and mental health traits. Nat. Genet. 51 , 1637–1644 (2019).

Zhao, B. et al. Large-scale GWAS reveals genetic architecture of brain white matter microstructure and genetic overlap with cognitive and mental health traits ( n = 17,706). Mol. Psychiatry https://doi.org/10.1038/s41380-019-0569-z (2019).

Lasagna-Reeves, C. A. et al. Reduction of Nuak1 decreases tau and reverses phenotypes in a tauopathy mouse model. Neuron 92 , 407–418 (2016).

Elliott, L. T. et al. Genome-wide association studies of brain imaging phenotypes in UK Biobank. Nature 562 , 210–216 (2018).

Smith, S. M. et al. An expanded set of genome-wide association studies of brain imaging phenotypes in UK Biobank. Nat. Neurosci. 24 , 737–745 (2021).

Lee, J. J. et al. Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals. Nat. Genet. 50 , 1112–1121 (2018).

Teslovich, T. M. et al. Biological, clinical and population relevance of 95 loci for blood lipids. Nature 466 , 707–713 (2010).

Sanchez-Roige, S. et al. Genome-Wide Association Study Meta-Analysis of the Alcohol Use Disorders Identification Test (AUDIT) in Two Population-Based Cohorts. Am. J. Psychiatry 176 , 107–118 (2019).

Locke, A. E. et al. Genetic studies of body mass index yield new insights for obesity biology. Nature 518 , 197–206 (2015).

Dashti, H. S. et al. Genome-wide association study identifies genetic loci for self-reported habitual sleep duration supported by accelerometer-derived estimates. Nat. Commun. 10 , 1100 (2019).

International Consortium for Blood Pressure Genome-Wide Association, S. et al. Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk. Nature 478 , 103–109 (2011).

Article   ADS   Google Scholar  

Johansson, A. et al. Genome-wide association and Mendelian randomization study of NT-proBNP in patients with acute coronary syndrome. Hum. Mol. Genet. 25 , 1447–1456 (2016).

Poorkaj, P. et al. Tau is a candidate gene for chromosome 17 frontotemporal dementia. Ann. Neurol. 43 , 815–825 (1998).

Baker, M. et al. Association of an extended haplotype in the tau gene with progressive supranuclear palsy. Hum. Mol. Genet. 8 , 711–715 (1999).

Johnson, N. C. XG: the forgotten blood group system. Immunohematology 27 , 68–71 (2011).

Kong, A. et al. The nature of nurture: Effects of parental genotypes. Science 359 , 424–428 (2018).

Zenaro, E., Piacentino, G. & Constantin, G. The blood-brain barrier in Alzheimer’s disease. Neurobiol. Dis. 107 , 41–56 (2017).

Pong, S., Karmacharya, R., Sofman, M., Bishop, J. R. & Lizano, P. The role of brain microvascular endothelial cell and blood-brain barrier dysfunction in schizophrenia. Complex Psychiatry 6 , 30–46 (2020).

Chatterjee, S. et al. Type 2 diabetes as a risk factor for dementia in women compared with men: a pooled analysis of 2.3 million people comprising more than 100,000 cases of dementia. Diabetes Care 39 , 300–307 (2016).

Veldsman, M. et al. Cerebrovascular risk factors impact frontoparietal network integrity and executive function in healthy ageing. Nat. Commun. 11 , 4340 (2020).

Power, M. C., Adar, S. D., Yanosky, J. D. & Weuve, J. Exposure to air pollution as a potential contributor to cognitive function, cognitive decline, brain imaging, and dementia: a systematic review of epidemiologic research. Neurotoxicology 56 , 235–253 (2016).

Grande, G., Ljungman, P. L. S., Eneroth, K., Bellander, T. & Rizzuto, D. Association between cardiovascular disease and long-term exposure to air pollution with the risk of dementia. JAMA Neurol. 77 , 801–809, (2020).

Ronnlund, M., Nyberg, L., Backman, L. & Nilsson, L. G. Stability, growth, and decline in adult life span development of declarative memory: cross-sectional and longitudinal data from a population-based study. Psychol. Aging 20 , 3–18 (2005).

Vidal-Pineiro, D. et al. Individual variations in ‘brain age’ relate to early-life factors more than to longitudinal brain change. Elife https://doi.org/10.7554/eLife.69995 (2021).

Fry, A. et al. Comparison of sociodemographic and health-related characteristics of UK Biobank participants with those of the general population. Am. J. Epidemiol. 186 , 1026–1034 (2017).

Alfaro-Almagro, F. et al. Image processing and Quality Control for the first 10,000 brain imaging datasets from UK Biobank. NeuroImage 166 , 400–424 (2018).

Miller, K. L. et al. Multimodal population brain imaging in the UK Biobank prospective epidemiological study. Nat. Neurosci. 19 , 1523–1536 (2016).

Good, C. D. et al. A voxel-based morphometric study of ageing in 465 normal adult human brains. NeuroImage 14 , 21–36 (2001).

Douaud, G. et al. Anatomically related grey and white matter abnormalities in adolescent-onset schizophrenia. Brain 130 , 2375–2386 (2007).

Groves, A. R. et al. Benefits of multi-modal fusion analysis on a large-scale dataset: life-span patterns of inter-subject variability in cortical morphometry and white matter microstructure. NeuroImage 63 , 365–380 (2012).

Smith, S. et al. Structural variability in the human brain reflects fine-grained functional architecture at the population level. J. Neurosci. 39 , 6136–6149 (2019).

Bycroft, C. et al. The UK Biobank resource with deep phenotyping and genomic data. Nature 562 , 203–209 (2018).

Mignogna, G. et al. Patterns of item nonresponse behaviour to survey questionnaires are systematic and associated with genetic loci. Nat. Hum. Behav . https://doi.org/10.1038/s41562-023-01632-7 (2023).

Hormozdiari, F., Kostem, E., Kang, E. Y., Pasaniuc, B. & Eskin, E. Identifying causal variants at loci with multiple signals of association. Genetics 198 , 497–508 (2014).

Thomas, P. D. et al. PANTHER: Making genome-scale phylogenetics accessible to all. Protein Sci. 31 , 8–22 (2022).

Ashburner, M. et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat. Genet. 25 , 25–29 (2000).

Gene Ontology, C. et al. The Gene Ontology knowledgebase in 2023. Genetics https://doi.org/10.1093/genetics/iyad031 (2023).

Imai, K., Keele, L. & Tingley, D. A general approach to causal mediation analysis. Psychol. Methods 15 , 309–334 (2010).

Hardy, J., de Strooper, B. & Escott-Price, V. Diabetes and Alzheimer’s disease: shared genetic susceptibility? Lancet Neurol. 21 , 962–964 (2022).

Carvalho-Silva, D. et al. Open Targets Platform: new developments and updates two years on. Nucleic Acids Res. 47 , D1056–D1065 (2019).

Gazal, S., Marquez-Luna, C., Finucane, H. K. & Price, A. L. Reconciling S-LDSC and LDAK functional enrichment estimates. Nat. Genet. 51 , 1202–1204 (2019).

Genomes Project, C. et al. A global reference for human genetic variation. Nature 526 , 68–74 (2015).

Jansen, I. E. et al. Author Correction: Genome-wide meta-analysis identifies new loci and functional pathways influencing Alzheimer’s disease risk. Nat. Genet. 52 , 354 (2020).

Trubetskoy, V. et al. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature 604 , 502–508 (2022).

Laird, A. R. et al. ALE meta-analysis workflows via the brainmap database: progress towards a probabilistic functional brain atlas. Front. Neuroinform. 3 , 23 (2009).

Download references

Acknowledgements

We are grateful to Profs Christian K. Tamnes, Lars T. Westlye, Kristine B. Walhovd and Anders M. Fjell, and Dr Andreas Engvig for providing the lifespan cohort which was used to initially derive the original ‘last in, first out’ brain network map, and to Prof Augustine Kong for helpful discussion on the associations between the PAR hit and early life and environmental factors. G.D. was supported by a UK MRC Career Development Fellowship (MR/K006673/1) and a Wellcome Collaborative Award (215573/Z/19/Z). S.S. was supported by Wellcome (203139/Z/16/Z; 215573/Z/19/Z). L.E. was funded by NSERC grants (RGPIN/05484-2019; DGECR/00118-2019) and a Michael Smith Health Research BC Scholar Award. A.M.W. received support through the NIH Intramural Research Program (ZIA-MH002781; ZIA-MH002782). This research was funded in whole, or in part, by the Wellcome Trust (215573/Z/19/Z; 203139/Z/16/Z; 203139/A/16/Z). For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. This research was also supported by the NIHR Oxford Health Biomedical Research Centre (NIHR203316). The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. The Wellcome Centre for Integrative Neuroimaging is supported by core funding from the Wellcome Trust (203139/Z/16/Z and 203139/A/16/Z).

Author information

These authors contributed equally: Lloyd T. Elliott, Anderson M. Winkler.

Authors and Affiliations

FMRIB Centre, Wellcome Centre for Integrative Neuroimaging (WIN), Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, UK

Jordi Manuello, Paul McCarthy, Fidel Alfaro-Almagro, Soojin Lee, Stephen Smith & Gwenaëlle Douaud

FOCUS Lab, Department of Psychology, University of Turin, Turin, Italy

Jordi Manuello

Department of Statistics and Actuarial Science, Simon Fraser University, Burnaby, Canada

Joosung Min & Lloyd T. Elliott

Pacific Parkinson’s Research Centre, The University of British Columbia, Vancouver, BC, Canada

National Institutes of Mental Health, National Institutes of Health, Bethesda, MD, USA

Anderson M. Winkler

Department of Human Genetics, University of Texas Rio Grande Valley, Brownsville, TX, USA

You can also search for this author in PubMed   Google Scholar

Contributions

G.D. conceived and supervised the work, and carried out some of the genetic and modifiable risk factors analyses. J.Ma. carried out most of the genetic and modifiable risk factors analyses. J.Mi., S.L., A.M.W., and L.T.E. carried out additional genetics analyses. G.D., P. McC., F.A.-A., S.S., and L.T.E. created/extracted the imaging and genetics data, and organised the non-imaging data and confound variables. L.T.E. co-supervised the genetic analyses. A.M.W. co-supervised the modifiable risk factor analyses. G.D. interpreted the results and wrote the paper. J.Ma., S.S., L.T.E., and A.M.W. revised the paper.

Corresponding author

Correspondence to Gwenaëlle Douaud .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Nature Communications thanks Xavier Caseras and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, peer review file, description of additional supplementary files, supplementary data 1-5, reporting summary, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Manuello, J., Min, J., McCarthy, P. et al. The effects of genetic and modifiable risk factors on brain regions vulnerable to ageing and disease. Nat Commun 15 , 2576 (2024). https://doi.org/10.1038/s41467-024-46344-2

Download citation

Received : 15 February 2023

Accepted : 22 February 2024

Published : 27 March 2024

DOI : https://doi.org/10.1038/s41467-024-46344-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

• Get Press Releases

• Media Contacts
• News Releases
• Photos & B-Roll Downloads
• VUMC Facts and Figures

Explore by Highlight

• Education & Training
• Growth & Finance
• Leadership Perspectives
• VUMC People

Explore by Topic

• Emergency & Trauma
• Genetics & Genomics
• Health Equity
• Health Policy
• Patient Spotlight
• Tech & Health

Explore by Location

• Monroe Carell Jr. Children’s Hospital at Vanderbilt
• Vanderbilt Bedford County Hospital
• Vanderbilt Health One Hundred Oaks
• Vanderbilt Health Affiliated Network
• Vanderbilt-Ingram Cancer Center
• Vanderbilt Psychiatric Hospital
• Vanderbilt Wilson County Hospital
• Vanderbilt Stallworth Rehabilitation Hospital
• Vanderbilt Tullahoma Harton Hospital
• Vanderbilt University Hospital

• Photos & B-Roll Downloads

Featured Story

VUMC performs its first combined lung and liver transplant

March 27, 2024, higher genetic risk of obesity means working out harder for same results .

Study authors used activity, clinical and genetic data from the National Institutes of Health’s All of Us Research Program to explore the association of genetic risk of higher body mass index and the level of physical activity needed to reduce incident obesity. 

Persons with a higher genetic risk of obesity need to work out harder than those of moderate or low genetic risk to avoid becoming obese, according to a Vanderbilt University Medical Center (VUMC) paper published in JAMA Network Open .  

Study authors used activity, clinical and genetic data from the National Institutes of Health’s All of Us Research Program to explore the association of genetic risk of higher body mass index and the level of physical activity needed to reduce incident obesity. 

“Physical activity guidelines do not account for individual differences,” said senior author Douglas Ruderfer , PhD, associate professor of Medicine, Division of Genetic Medicine, and director of the Center for Digital Genomic Medicine at VUMC. “Genetic background contributes to the amount of physical activity needed to mitigate obesity. The higher the genetic risk, the more steps needed per day.  

“I think an important component to this result is that individuals can be active enough to account for their genetic background, or their genetic risk for obesity, regardless of how high that risk might be,” he added. “And there are many other contributors that play a role including diet and environmental factors.” 

Included in the study were 3,124 middle-aged participants without obesity who owned a Fitbit device and walked an average of 8,326 steps per day for a median of more than 5 years. The incidence of obesity over the study period increased from 13% to 43% in the lowest and highest polygenic risk score groups.  

Most importantly, I would like for patients to know that your genetic risk doesn’t determine your overall risk of obesity, and you can actually overcome that risk by being more active Lead author Evan Brittain , MD

Individuals with a polygenic risk score in the 75 th percentile would need to walk an average of 2,280 more steps per day (a total of 11,020 steps per day) than those in the 50 th percentile to have a comparable risk of obesity, according to the study.  

Persons with a baseline BMI of 22, 24, 26 and 28 who were in the 75 th percentile of polygenic risk score would need to walk an additional 3,460, 4,430, 5,380 and 6,350 steps per day, respectively, to have a comparable risk of obesity to persons in the 25 th percentile. 

“I think it is intuitive that individuals who have a higher genetic risk of obesity might need to have more physical activity to reduce that risk, but what is new and important from this study is that we were able to put a number on the amount of activity needed to reduce the risk,” said lead author Evan Brittain , MD, associate professor of Medicine in the Division of Cardiovascular Medicine at VUMC and lead investigator in Digital Health for the All of Us Research Program Data and Research Center. 

“It is becoming more commonplace to know you have a genetic risk for obesity in the genomic era when genetic results are being returned directly to patients. And you can imagine a future in which that data could be integrated with someone’s electronic health record and could form the basis of an individual’s physical activity recommendation from their doctor.” 

Study authors said they now want to see if the findings generalize to more representative and diverse populations in order to determine if providing information for individual activity recommendations results in improved health and a lower likelihood of obesity. 

“We would like to test whether knowledge of one’s genetic risk for obesity actually has an impact on their behavior,” Brittain said. “I think these findings could be empowering for patients because the current physical activity guidelines take a one-size-fits-all approach, and what we learned is that depending on your genetic risk, the guidelines may underestimate the amount of activity needed to reduce your risk of obesity. 

“Most importantly, I would like for patients to know that your genetic risk doesn’t determine your overall risk of obesity, and you can actually overcome that risk by being more active,” he added.

To access broadcast quality video for download, fill out the submission form.

Related Articles

November 19, 2019

Getting the goods on obesity.

Obesity and two post-operative complications linked with it have associated genetic variants in common, suggesting that obesity may be the culprit.

By Paul Govern

January 20, 2012

Obesity genes linked to uterine cancer.

In addition to body mass index, genetic markers of obesity may provide value in predicting endometrial cancer risk.

By Vumc News And Communications

February 1, 2024

A biomarker for early type 2 diabetes.

Genetic analyses suggest that branched chain amino acids may be a sensitive biomarker of early or subclinical Type 2 diabetes and could be used to identify risk and implement preventive measures.

By Bill Snyder