Inbreeding depression and the evolution of dispersal rates: a multilocus model

Inbreeding depression is one of the possible reasons organisms disperse. In this article, we present a two-locus model for the evolution of dispersal in the presence of inbreeding depression. The first locus codes for a modifier of the migration rate, while the second locus is a selected locus generating inbreeding depression. We express the change in frequency of the migration modifier as a function of allele frequencies and genetic associations and then use a quasi-equilibrium assumption to express genetic associations as functions of allele frequencies. Our model disentangles two effects of inbreeding depression: it gives an advantage to migrant individuals because their offspring are on average less homozygous, but it also decreases the degree of population structure, thus decreasing the strength of kin selection for dispersal. We then extend our model to include an infinite number of selected loci. When the cost of dispersal is not too high, the model predictions are confirmed by multilocus simulation results and show that inbreeding depression can have a substantial effect on the dispersal rate. For high costs of dispersal, we observe discrepancies between the model and the simulations, probably caused by associations among selected loci, which are neglected in the analysis.     

Molecular‐clock methods for estimating evolutionary rates and timescales

The molecular clock presents a means of estimating evolutionary rates and timescales using genetic data. These estimates can lead to important insights into evolutionary processes and mechanisms, as well as providing a framework for further biological analyses. To deal with rate variation among genes and among lineages, a diverse range of molecular‐clock methods have been developed. These methods have been implemented in various software packages and differ in their statistical properties, ability to handle different models of rate variation, capacity to incorporate various forms of calibrating information and tractability for analysing large data sets. Choosing a suitable molecular‐clock model can be a challenging exercise, but a number of model‐selection techniques are available. In this review, we describe the different forms of evolutionary rate heterogeneity and explain how they can be accommodated in molecular‐clock analyses. We provide an outline of the various clock methods and models that are available, including the strict clock, local clocks, discrete clocks and relaxed clocks. Techniques for calibration and clock‐model selection are also described, along with methods for handling multilocus data sets. We conclude our review with some comments about the future of molecular clocks.     

The relation between multilocus population genetics and social evolution theory

Evolution at multiple gene positions is complicated. Direct selection on one gene disturbs the evolutionary dynamics of associated genes. Recent years have seen the development of a multilocus methodology for modeling evolution at arbitrary numbers of gene positions with arbitrary dominance and epistatic relations, mode of inheritance, genetic linkage, and recombination. We show that the approach is conceptually analogous to social evolutionary methodology, which focuses on selection acting on associated individuals. In doing so, we (1) make explicit the links between the multilocus methodology and the foundations of social evolution theory, namely, Price's theorem and Hamilton's rule; (2) relate the multilocus approach to levels-of-selection and neighbor-modulated-fitness approaches in social evolution; (3) highlight the equivalence between genetical hitchhiking and kin selection; (4) demonstrate that the multilocus methodology allows for social evolutionary analyses involving coevolution of multiple traits and genetical associations between nonrelatives, including individuals of different species; (5) show that this methodology helps solve problems of dynamic sufficiency in social evolution theory; (6) form links between invasion criteria in multilocus systems and Hamilton's rule of kin selection; (7) illustrate the generality and exactness of Hamilton's rule, which has previously been described as an approximate, heuristic result.     

Multilocus selection in subdivided populations II. Maintenance of polymorphism under weak or strong migration

The potential of maintaining multilocus polymorphism by migration-selection balance is studied. A large population of diploid individuals is distributed over finitely many demes connected by migration. Generations are discrete and nonoverlapping, selection may vary across demes, and loci are multiallelic. It is shown that if migration and recombination are strong relative to selection, then with weak or no epistasis and intermediate dominance at every locus and in every deme, arbitrarily many alleles can be maintained at arbitrarily many loci at a stable equilibrium. If migration is weak relative to selection and recombination, then with weak or no epistasis and intermediate dominance at every locus and in every deme, as many alleles as there are demes can be maintained at arbitrarily many loci at equilibrium. In both cases open sets of such parameter combinations are constructed, thus the results are robust with respect to small, but arbitrary, perturbations in the parameters. For weak migration, the number of demes is, in fact, a generic upper bound to the number of alleles that can be maintained at any locus. Thus, several scenarios are identified under which multilocus polymorphism can be maintained by migration-selection balance when this is impossible in a panmictic population.      

Associations between cytoplasmic and nuclear Loci in hybridizing populations

We extend current multilocus models to describe the effects of migration, recombination, selection, and nonrandom mating on sets of genes in diploids with varied modes of inheritance, allowing us to consider the patterns of nuclear and cytonuclear associations (disequilibria) under various models of migration. We show the relationship between the multilocus notation recently presented by Kirkpatrick, Johnson, and Barton (developed from previous work by Barton and Turelli) and the cytonuclear parameterization of Asmussen, Arnold, and Avise and extend this notation to describe associations between cytoplasmic elements and multiple nuclear genes. Under models with sexual symmetry, both nuclear-nuclear and cytonuclear disequilibria are equivalent. They differ, however, in cases involving some type of sexual asymmetry, which is then reflected in the asymmetric inheritance of cytoplasmic markers. An example given is the case of different migration rates in males and females; simulations using 2, 3, 4, or 5 unlinked autosomal markers with a maternally inherited cytoplasmic marker illustrate how nuclear-nuclear and cytonuclear associations can be used to separately estimate female and male migration rates. The general framework developed here allows us to investigate conditions where associations between loci with different modes of inheritance are not equivalent and to use this nonequivalence to test for deviations from simple models of admixture.     

Conditions for sympatric speciation: A diploid model incorporating habitat fidelity and non-habitat assortative mating

Summary Three types of genes have been proposed to promote sympatric speciation: habitat preference genes, assortative mating genes and habitat-based fitness genes. Previous computer models have analysed these genes separately or in pairs. In this paper we describe a multilocus model in which genes of all three types are considered simultaneously. Our computer simulations show that speciation occurs in complete sympatry under a broad range of conditions. The process includes an initial diversification phase during which a slight amount of divergence occurs, a quasi-equilibrium phase of stasis during which little or no detectable divergence occurs and a completion phase during which divergence is dramatic and gene flow between diverging habitat morphs is rapidly eliminated. Habitat preference genes and habitat-specific fitness genes become associated when assortative mating occurs due to habitat preference, but interbreeding between individuals adapted to different habitats occurs unless habitat preference is almost error free. However, nonhabitat assortative mating, when coupled with habitat preference can eliminate this interbreeding. Even when several loci contribute to the probability of expression of non-habitat assortative mating and the contributions of individual loci are small, gene flow between diverging portions of the population can terminate within less than 1000 generations.     

EPISTASIS AND THE INCREASE IN ADDITIVE GENETIC VARIANCE: IMPLICATIONS FOR PHASE 1 OF WRIGHT'S SHIFTING-BALANCE PROCESS

Central to Wright's shifting-balance theory is the idea that genetic drift and selection in systems with gene interaction can lead to the formation of “adaptive gene complexes.” The theory of genetic drift has been well developed over the last 60 years; however, nearly all of this theory is based on the assumption that only additive gene effects are acting. Wright's theory was developed recognizing that there was a “universality of interaction effects,” which implies that additive theory may not be adequate to describe the process of differentiation that Wright was considering. The concept of an adaptive gene complex implies that an allele that is favored by individual selection in one deme may be removed by selection in another deme. In quantitative genetic terms, the average effects of an allele relative to other alleles changes from deme to deme. The model presented here examines the variance in local breeding values (LBVs) of a single individual and the covariance in the LBVs of a pair of individuals mated in the same deme relative to when they are mated in different demes. Local breeding value is a measure of the average effects of the alleles that make up that individual in a particular deme. I show that when there are only additive effects the covariance between the LBVs of individuals equals the variance in the LBV of an individual. As the amount of epistasis in the ancestral population increases, the variance in the LBV of an individual increases and the covariance between the LBVs of a pair of individuals decreases. The divergence in these two values is a measure of the extent to which the LBV of an individual varies independently of the LBVs of other individuals. When this value is large, it means that the relative ordering of the average effects of alleles will change from deme to deme. These results confirm an important component of Wright's shifting-balance theory: When there is gene interaction, genetic drift can lead to the reordering of the average effects of alleles and when coupled with selection this will lead to the formation of the adaptive gene complexes.     

The stability of symmetric solutions to polygenic models

Analysis of multilocus evolution is usually intractable for more than n approximately 10 genes, because the frequencies of very large numbers of genotypes must be followed. An exact analysis of up to n approximately 100 loci is feasible for a symmetrical model, in which a set of unlinked loci segregate for two alleles (labeled "0" and "1") with interchangeable effects on fitness. All haploid genotypes with the same number of 1 alleles can then remain equally frequent. However, such a symmetrical solution may be unstable: for example, under stabilizing selection, populations tend to fix any one genotype which approaches the optimum. Here, we show how the 2(n)x2(n) stability matrix can be decomposed into a set of matrices, each no larger than nxn. This allows the stability of symmetrical solutions to be determined. We apply the method to stabilizing and disruptive selection in a single deme and to selection against heterozygotes in a linear cline.     

Multilocus analysis of hypertension: a hierarchical approach

While hypertension is a complex disease with a well-documented genetic component, genetic studies often fail to replicate findings. One possibility for such inconsistency is that the underlying genetics of hypertension is not based on single genes of major effect, but on interactions among genes. To test this hypothesis, we studied both single locus and multilocus effects, using a case-control design of subjects from Ghana. Thirteen polymorphisms in eight candidate genes were studied. Each candidate gene has been shown to play a physiological role in blood pressure regulation and affects one of four pathways that modulate blood pressure: vasoconstriction (angiotensinogen, angiotensin converting enzyme - ACE, angiotensin II receptor), nitric oxide (NO) dependent and NO independent vasodilation pathways and sodium balance (G protein-coupled receptor kinase, GRK4). We evaluated single site allelic and genotypic associations, multilocus genotype equilibrium and multilocus genotype associations, using multifactor dimensionality reduction (MDR). For MDR, we performed systematic reanalysis of the data to address the role of various physiological pathways. We found no significant single site associations, but the hypertensive class deviated significantly from genotype equilibrium in more than 25% of all multilocus comparisons (2,162 of 8,178), whereas the normotensive class rarely did (11 of 8,178). The MDR analysis identified a two-locus model including ACE and GRK4 that successfully predicted blood pressure phenotype 70.5% of the time. Thus, our data indicate epistatic interactions play a major role in hypertension susceptibility. Our data also support a model where multiple pathways need to be affected in order to predispose to hypertension.     

A Bayesian approach to the evolution of social learning

There has been much interest in understanding the evolution of social learning. Investigators have tried to understand when natural selection will favor individuals who imitate others, how imitators should deal with the fact that available models may exhibit different behaviors, and how social and individual learning should interact. In all of this work, social learning and individual learning have been treated as alternative, conceptually distinct processes. Here we present a Bayesian model in which both individual and social learning arise from a single inferential process. Individuals use Bayesian inference to combine social and nonsocial cues about the current state of the environment. This model indicates that natural selection favors individuals who place heavy weight on social cues when the environment changes slowly or when its state cannot be well predicted using nonsocial cues. It also indicates that a conformist bias should be a universal aspect of social learning.     

EVOLUTION OF ALTRUISM IN KIN-STRUCTURED AND RANDOM SUBDIVIDED POPULATIONS

A. population structure favorable to the evolution of an altruistic trait is studied by Monte Carlo simulation. The model is based on a small-scale nonindustrial human society but seems generalizable to other highly social mammals. Three hierarchical levels are recognized: 1) the ecologically isolated local group (hamlet) which may be composed of kin and/or unrelated individuals; 2) the deme (settlement) comprising several such groups which interbreed; and 3) the set of demes (metapopulation) among which gene flow occurs. The first two levels of the model are based on D. S. Wilson's structured deme concept; the third allows for gene flow among demes in the metapopulation and for the structured diffusion of alleles across a wider area than might be included within the scope of a single deme. The simulation models genetic drift by a process of hamlet formation which may be random, or variously kin-structured. Hamlets may then become extinct based on a probability function of their gene frequencies. Individual selection within settlements is modeled deterministically, and gene flow among settlements is modeled as two-dimensional steppingstone migration of random or kin-structured groups. Results of the simulations show that, with realistic values for group sizes, moderate extinction rate, and high rates of migration (m > 27%), disadvantageous alleles (s = 10% and 25%) may increase markedly due to differential hamlet extinction over the course of 50 generations. The greater the degree of kin-structuring of founder groups, the higher the variance among hamlets and the faster the rate of increase of the allele for altruism. Nonetheless, even in some randomly founded groups, a clear increase in the altruism gene frequency occurred. It is also notable that kin-structured group selection by hamlet extinction may be effective when the initial frequency of altruism genes is very low (average of one per deme) and among a relatively small number of demes (25). Thus the process of group extinction in a hierarchically structured population allows rapid increase of an allele for altruism under plausible demographic conditions.     

Modelling multilocus selection in an individual‐based,spatially‐explicit landscape genetics framework

We implemented multilocus selection in a spatially‐explicit, individual‐based framework that enables multivariate environmental gradients to drive selection in many loci as a new module for the landscape genetics programs, CDPOP and CDMetaPOP. Our module simulates multilocus selection using a linear additive model, providing a flexible platform to evaluate a wide range of genotype‐environment associations. Importantly, the module allows simulation of selection in any number of loci under the influence of any number of environmental variables. We validated the module with individual‐based selection simulations under Wright‐Fisher assumptions. We then evaluated results for simulations under a simple landscape selection model. Next, we simulated individual‐based multilocus selection across a complex selection landscape with three loci linked to three different environmental variables. Finally, we demonstrated how the program can be used to simulate multilocus selection under varying selection strengths across different levels of gene flow in a landscape genetics framework. This new module provides a valuable addition to the study of landscape genetics, allowing for explicit evaluation of the contributions and interactions between gene flow and selection‐driven processes across complex, multivariate environmental and landscape conditions.     

Genes@Work: an efficient algorithm for pattern discovery and multivariate feature selection in gene expression data

MOTIVATION: Despite the growing literature devoted to finding differentially expressed genes in assays probing different tissues types, little attention has been paid to the combinatorial nature of feature selection inherent to large, high-dimensional gene expression datasets. New flexible data analysis approaches capable of searching relevant subgroups of genes and experiments are needed to understand multivariate associations of gene expression patterns with observed phenotypes. RESULTS: We present in detail a deterministic algorithm to discover patterns of multivariate gene associations in gene expression data. The patterns discovered are differential with respect to a control dataset. The algorithm is exhaustive and efficient, reporting all existent patterns that fit a given input parameter set while avoiding enumeration of the entire pattern space. The value of the pattern discovery approach is demonstrated by finding a set of genes that differentiate between two types of lymphoma. Moreover, these genes are found to behave consistently in an independent dataset produced in a different laboratory using different arrays, thus validating the genes selected using our algorithm. We show that the genes deemed significant in terms of their multivariate statistics will be missed using other methods. AVAILABILITY: Our set of pattern discovery algorithms including a user interface is distributed as a package called Genes@Work. This package is freely available to non-commercial users and can be downloaded from our website (http://www.research.ibm.com/FunGen).     

The interpretation of selection coefficients

Evolutionary biologists have an array of powerful theoretical techniques that can accurately predict changes in the genetic composition of populations. Changes in gene frequencies and genetic associations between loci can be tracked as they respond to a wide variety of evolutionary forces. However, it is often less clear how to decompose these various forces into components that accurately reflect the underlying biology. Here, we present several issues that arise in the definition and interpretation of selection and selection coefficients, focusing on insights gained through the examination of selection coefficients in multilocus notation. Using this notation, we discuss how its flexibility—which allows different biological units to be identified as targets of selection—is reflected in the interpretation of the coefficients that the notation generates. In many situations, it can be difficult to agree on whether loci can be considered to be under “direct” versus “indirect” selection, or to quantify this selection. We present arguments for what the terms direct and indirect selection might best encompass, considering a range of issues, from viability and sexual selection to kin selection. We show how multilocus notation can discriminate between direct and indirect selection, and describe when it can do so.     

Selection versus random drift: long-term polymorphism persistence in small populations (evidence and modelling)

Our data on a subterranean mammal, Spalax ehrenbergi, and other evidence, indicate that appreciable polymorphism can be preserved in small isolated populations consisting of several dozens of, or a hundred, individuals. Current theoretical models predict fast gene fixation in small panmictic populations without selection, mutation, or gene inflow. Using simple multilocus models, we demonstrate here that moderate stabilizing selection (with stable or fluctuating optimum) for traits controlled by additive genes could oppose random fixation in such isolates during thousands of generations. We also show that in selection-free models polymorphism persists only for a few hundred generations even under high mutation rates. Our multi-chromosome models challenge the hitchhiking hypothesis of polymorphism maintenance for many neutral loci due to close linkage with few selected loci.     

COMPARISON OF NON‐GAUSSIAN QUANTITATIVE GENETIC MODELS FOR MIGRATION AND STABILIZING SELECTION

The balance between stabilizing selection and migration of maladapted individuals has formerly been modeled using a variety of quantitative genetic models of increasing complexity, including models based on a constant expressed genetic variance and models based on normality. The infinitesimal model can accommodate nonnormality and a nonconstant genetic variance as a result of linkage disequilibrium. It can be seen as a parsimonious one‐parameter model that approximates the underlying genetic details well when a large number of loci are involved. Here, the performance of this model is compared to several more realistic explicit multilocus models, with either two, several or a large number of alleles per locus with unequal effect sizes. Predictions for the deviation of the population mean from the optimum are highly similar across the different models, so that the non‐Gaussian infinitesimal model forms a good approximation. It does, however, generally estimate a higher genetic variance than the multilocus models, with the difference decreasing with an increasing number of loci. The difference between multilocus models depends more strongly on the effective number of loci, accounting for relative contributions of loci to the variance, than on the number of alleles per locus.     

Determination of Wolbachia diversity in butterflies from Western Ghats, India, by a multigene approach

Members of the genus Wolbachia are intracellular bacteria that are widespread in arthropods and establish diverse symbiotic associations with their hosts, ranging from mutualism to parasitism. Here we present the first detailed analyses of Wolbachia in butterflies from India with screening of 56 species. Twenty-nine species (52%) representing five families were positive for Wolbachia. This is the first report of Wolbachia infection in 27 of the 29 species; the other two were reported previously. This study also provides the first evidence of infection in the family Papilionidae. A striking diversity was observed among Wolbachia strains in butterfly hosts based on five multilocus sequence typing (MLST) genes, with 15 different sequence types (STs). Thirteen STs are new to the MLST database, whereas ST41 and ST125 were reported earlier. Some of the same host species from this study carried distinctly different Wolbachia strains, whereas the same or different butterfly hosts also harbored closely related Wolbachia strains. Butterfly-associated STs in the Indian sample originated by recombination and point mutation, further supporting the role of both processes in generating Wolbachia diversity. Recombination was detected only among the STs in this study and not in those from the MLST database. Most of the strains were remarkably similar in their wsp genotype, despite divergence in MLST. Only two wsp alleles were found among 25 individuals with complete hypervariable region (HVR) peptide profiles. Although both wsp and MLST show variability, MLST gives better separation between the strains. Completely different STs were characterized for the individuals sharing the same wsp alleles.     

Zygotic associations and multilocus statistics in a nonequilibrium diploid population

The usual approach to characterizing and estimating multilocus associations in a diploid population assumes that the population is in Hardy-Weinberg equilibrium. The purpose of this study is to develop a set of summary statistics that can be used to characterize and estimate the multilocus associations in a nonequilibrium population. The concept of "zygotic associations" is first expanded to facilitate the development. The summary statistics are calculated using the distribution of a random variable, the number of heterozygous loci (K) found in diploid individuals in the population. In particular, the variance of K consists of single-locus and multilocus components with the latter being the sum of zygotic associations between pairs of loci. Simulation results show that the multilocus associations in the variance of K are detectable in a sample of moderate size (> or =30) when the sum of all pairwise zygotic associations is greater than zero and when gene frequency is intermediate. The method presented here is a generalization of the well-known development for the Hardy-Weinberg equilibrium population and thus may be of more general use in elucidating the multilocus organizations in nonequilibrium and equilibrium populations.     

Selection and drift in subdivided populations: a straightforward method for deriving diffusion approximations and applications involving dominance, selfing and local extinctions

Population structure affects the relative influence of selection and drift on the change in allele frequencies. Several models have been proposed recently, using diffusion approximations to calculate fixation probabilities, fixation times, and equilibrium properties of subdivided populations. We propose here a simple method to construct diffusion approximations in structured populations; it relies on general expressions for the expectation and variance in allele frequency change over one generation, in terms of partial derivatives of a "fitness function" and probabilities of genetic identity evaluated in a neutral model. In the limit of a very large number of demes, these probabilities can be expressed as functions of average allele frequencies in the metapopulation, provided that coalescence occurs on two different timescales, which is the case in the island model. We then use the method to derive expressions for the probability of fixation of new mutations, as a function of their dominance coefficient, the rate of partial selfing, and the rate of deme extinction. We obtain more precise approximations than those derived by recent work, in particular (but not only) when deme sizes are small. Comparisons with simulations show that the method gives good results as long as migration is stronger than selection.     

Age-specific effect of heterozygosity on survival in alpine marmots, Marmota marmota

The fitness consequences of heterozygosity and the mechanisms underpinning them are still highly controversial. Using capture–mark–recapture models, we investigated the effects of individual heterozygosity, measured at 16 microsatellite markers, on age-dependent survival and access to dominance in a socially monogamous mammalian species, the alpine marmot. We found a positive correlation between standardized multilocus heterozygosity and juvenile survival. However, there was no correlation between standardized multilocus heterozygosity and either survival of older individuals or access to dominance. The disappearance of a significant heterozygosity fitness correlation when individuals older than juveniles are considered is consistent with the prediction that differences in survival among individuals are maximal early in life. The lack of a correlation between heterozygosity and access to dominance may be a consequence of few homozygous individuals attaining the age at which they might reach dominance. Two hypotheses have been proposed to explain heterozygosity-fitness correlations: genome-wide effects reflected by all markers or local effects of specific markers linked to genes that determine fitness. In accordance with genome-wide effects of heterozygosity, we found significant correlations between heterozygosities calculated across single locus or across two sets of eight loci. Thus, the genome-wide heterozygosity effect seems to explain the observed heterozygosity-fitness correlation in the alpine marmot.