C. A. Buerkle

Hybridization and speciation

Hybridization has many and varied impacts on the process of speciation. Hybridization may slow or reverse differentiation by allowing gene flow and recombination. It may accelerate speciation via adaptive introgression or cause near‐instantaneous speciation by allopolyploidization. It may have multiple effects at different stages and in different spatial contexts within a single speciation event. We offer a perspective on the context and evolutionary significance of hybridization during speciation, highlighting issues of current interest and debate. In secondary contact zones, it is uncertain if barriers to gene flow will be strengthened or broken down due to recombination and gene flow. Theory and empirical evidence suggest the latter is more likely, except within and around strongly selected genomic regions. Hybridization may contribute to speciation through the formation of new hybrid taxa, whereas introgression of a few loci may promote adaptive divergence and so facilitate speciation. Gene regulatory networks, epigenetic effects and the evolution of selfish genetic material in the genome suggest that the Dobzhansky–Muller model of hybrid incompatibilities requires a broader interpretation. Finally, although the incidence of reinforcement remains uncertain, this and other interactions in areas of sympatry may have knock‐on effects on speciation both within and outside regions of hybridization.

Variable patterns of introgression in two sculpin hybrid zones suggest that genomic isolation differs among populations

Theory predicts that reproductive isolation may be due to intrinsic genetic incompatibilities or extrinsic ecological factors. Therefore, an understanding of the genetic basis of isolation may require analyses of evolutionary processes in situ to include environmental factors. Here we study genetic isolation between populations of sculpins (Cottus) at 168 microsatellites. Genomic clines were fit using 480 individuals sampled across independent natural hybrid zones that have formed between one invading species and two separate populations of a resident species. Our analysis tests for deviations from neutral patterns of introgression at individual loci based on expectations given genome‐wide admixture. Roughly 51% of the loci analysed displayed significant deviations. An overall deficit of interspecific heterozygotes in 26% and 21% of the loci suggests that widespread underdominance drives genomic isolation. At the same time, selection promotes introgression of almost 30% of the markers, which implies that hybridization may increase the fitness of admixed individuals. Cases of overdominance or epistatic interactions were relatively rare. Despite the similarity of the two hybrid zones in their overall genomic composition, patterns observed at individual loci show little correlation between zones and many fit different genotypic models of fitness. At this point, it remains difficult to determine whether these results are due to differences in external selection pressures or cryptic genetic differentiation of distinct parental populations. In the future, data from mapped genetic markers and on variation of ecological factors will provide additional insights into the contribution of these factors to variation in the evolutionary consequences of hybridization.

The genomic consequences of adaptive divergence and reproductive isolation between species of manakins

The processes of adaptation and speciation are expected to shape genomic variation within and between diverging species. Here we analyze genomic heterogeneity of genetic differentiation and introgression in a hybrid zone between two bird species (Manacus candei and M. vitellinus) using 59 100 SNPs, a whole genome assembly, and Bayesian models. Measures of genetic differentiation ( FST ) and introgression (genomic cline center [α] and rate [β]) were highly heterogeneous among loci. We identified thousands of loci with elevated parameter estimates, some of which are likely to be associated with variation in fitness in Manacus populations. To analyze the genomic organization of differentiation and introgression, we mapped SNPs onto a draft assembly of the M. vitellinus genome. Estimates of FST , α, and β were autocorrelated at very short physical distances (< 100 bp), but much less so beyond this. In addition, average statistical associations (linkage disequilibrium) between SNPs were generally low and were not higher in admixed populations than in populations of the parental species. Although they did not occur with a constant probability across the genome, loci with elevated FST , α, and β were not strongly co‐localized in the genome. Contrary to verbal models that predict clustering of loci involved in adaptation and isolation in discrete genomic regions, these results are consistent with the hypothesis that genetic regions involved in adaptive divergence and reproductive isolation are scattered throughout the genome. We also found that many loci were characterized by both exceptional genetic differentiation and introgression, consistent with the hypothesis that loci involved in isolation are also often characterized by a history of divergent selection. However, the concordance between isolation and differentiation was only partial, indicating a complex architecture and history of loci involved in isolation.

Genome scan of hybridizing sunflowers from Texas (Helianthus annuus and H. debilis) reveals asymmetric patterns of introgression and small islands of genomic differentiation

Although the sexual transfer of genetic material between species (i.e. introgression) has been documented in many groups of plants and animals, genome‐wide patterns of introgression are poorly understood. Is most of the genome permeable to interspecific gene flow, or is introgression typically restricted to a handful of genomic regions? Here, we assess the genomic extent and direction of introgression between three sunflowers from the south‐central USA: the common sunflower, Helianthus annuus ssp. annuus; a near‐endemic to Texas, Helianthus debilis ssp. cucumerifolius; and their putative hybrid derivative, thought to have recently colonized Texas, H. annuus ssp. texanus. Analyses of variation at 88 genetically mapped microsatellite loci revealed that long‐term migration rates were high, genome‐wide and asymmetric, with higher migration rates from H. annuus texanus into the two parental taxa than vice versa. These results imply a longer history of intermittent contact between H. debilis and H. annuus than previously believed, and that H. annuus texanus may serve as a bridge for the transfer of alleles between its parental taxa. They also contradict recent theory suggesting that introgression should predominantly be in the direction of the colonizing species. As in previous studies of hybridizing sunflower species, regions of genetic differentiation appear small, whether estimated in terms of FST or unidirectional migration rates. Estimates of recent immigration and admixture were inconsistent, depending on the type of analysis. At the individual locus level, one marker showed striking asymmetry in migration rates, a pattern consistent with tight linkage to a Bateson–Dobzhansky–Muller incompatibility.

Experimental evidence for ecological selection on genome variation in the wild

Understanding natural selections effect on genetic variation is a major goal in biology, but the genome-scale consequences of contemporary selection are not well known. In a release and recapture field experiment we transplanted stick insects to native and novel host plants and directly measured allele frequency changes within a generation at 186 576 genetic loci. We observed substantial, genome-wide allele frequency changes during the experiment, most of which could be attributed to random mortality (genetic drift). However, we also documented that selection affected multiple genetic loci distributed across the genome, particularly in transplants to the novel host. Host-associated selection affecting the genome acted on both a known colour-pattern trait as well as other (unmeasured) phenotypes. We also found evidence that selection associated with elevation affected genome variation, although our experiment was not designed to test this. Our results illustrate how genomic data can identify previously underappreciated ecological sources and phenotypic targets of selection.

Recombinant hybrids retain heterozygosity at many loci: new insights into the genomics of reproductive isolation in Populus

The maintenance of species barriers in the face of gene flow is often thought to result from strong selection against intermediate genotypes, thereby preserving genetic differentiation. Most speciation genomic studies thus aim to identify exceptionally divergent loci between populations, but divergence will be affected by many processes other than reproductive isolation (RI) and speciation. Through genomic studies of recombinant hybrids sampled in the wild, genetic variation associated with RI can be observed in situ, because selection against incompatible genotypes will leave detectable patterns of variation in the hybrid genomes. To better understand the mechanisms directly involved in RI, we investigated three natural ‘replicate’ hybrid zones between two divergent Populus species via locus‐specific patterns of ancestry across recombinant hybrid genomes. As expected, genomic patterns in hybrids and their parental species were consistent with the presence of underdominant selection at several genomic regions. Surprisingly, many loci displayed greatly increased between‐species heterozygosity in recombinant hybrids despite striking genetic differentiation between the parental genomes, the opposite of what would be expected with selection against intermediate genotypes. Only a limited, reproducible set of genotypic combinations was present in hybrid genomes across localities. In the absence of clearly delimited ‘hybrid habitats’, our results suggest that complex epistatic interactions within genomes play an important role in advanced stages of RI between these ecologically divergent forest trees. This calls for more genomic studies that test for unusual patterns of genomic ancestry in hybridizing species.

bgc: Software for Bayesian estimation of genomic clines.

Introgression in admixed populations can be used to identify candidate loci that might underlie adaptation or reproductive isolation. The Bayesian genomic cline model provides a framework for quantifying variable introgression in admixed populations and identifying regions of the genome with extreme introgression that are potentially associated with variation in fitness. Here we describe the bgc software, which uses Markov chain Monte Carlo to estimate the joint posterior probability distribution of the parameters in the Bayesian genomic cline model and designate outlier loci. This software can be used with next‐generation sequence data, accounts for uncertainty in genotypic state, and can incorporate information from linked loci on a genetic map. Output from the analysis is written to an HDF5 file for efficient storage and manipulation. This software is written in C++. The source code, software manual, compilation instructions and example data sets are available under the GNU Public License at http://sites.google.com/site/bgcsoftware/.

The n = 1 constraint in population genomics.

A key objective of population genomics is to identify portions of the genome that have been shaped by natural selection rather than by neutral divergence. A previously recognized but underappreciated challenge to this objective is that observations of allele frequencies across genomes in natural populations often correspond to a single, unreplicated instance of the outcome of evolution. This is because the composition of each individual genomic region and population is expected to be the outcome of a unique array of evolutionary processes. Given a single observation, inference of the evolutionary processes that led to the observed state of a locus is associated with considerable uncertainty. This constraint on inference can be ameliorated by utilizing multi‐allelic (e.g. DNA haplotypes) rather than bi‐allelic markers, by analysing two or more populations with certain models and by utilizing studies of replicated experimental evolution. Future progress in population genomics will follow from research that recognizes the ‘n = 1 constraint’ and that utilizes appropriate and explicit evolutionary models for analysis.

Inconvenient truths in population and speciation genetics point towards a future beyond allele frequencies

In 1999 Whitlock & McCauley (1999) illustrated that FST does not have a simple relationship to migration (4 Nm). For example, in empirical studies of populations, we typically observe a single realization of a combination of processes that include stochasticity and genetic drift so that we do not expect FST to match expectations exactly (Buerkle et al., 2011). Scientists using approximate Bayesian computation to infer evolutionary histories from coalescent simulations have learned to test empirically which statistics calculated for simulated genetic data were correlated with, and therefore informative about, evolutionary parameters of the simulated history (Wegmann et al., 2009; Aeschbacher et al., 2012). In other words, even in simple simulations of populations with migration, we cannot readily predict which measures on the genome will record the differences in evolutionary histories that were simulated. Undeterred, in speciation and landscape genetics, we have often utilized measures of FST as if they were indicative of historical migration, when the reality is much more complex. Ravinet et al. (this volume) provide an extensive and thorough discussion of many of the relevant complexities. The cautionary notes include mutation and recombination, for which we rarely have relevant information in many empirical systems. And even when we have estimates, say for recombination, these are for a particular cross (e.g. an F2 linkage map from a single pair of individuals), or in the context of a particular population. It remains unknown to what extent these estimates predict recombination in wider crosses, such as those following secondary contact and relevant to speciation. Ravinet et al. come to a fairly optimistic conclusion that the obstacles are surmountable and we can make progress in finding barrier loci, and they offer suggestions for the way forward. If in future studies we are mindful of the cautionary notes of Ravinet et al. and others (e.g. Whitlock & McCauley, 1999), then I can share their optimism. For one, we will need to move beyond simply equating transformations of allele frequencies to evolutionary processes, including gene flow or its absence. Despite the extensive treatment of complexities, Ravinet et al. could have gone on and highlighted additional significant considerations that should heighten our scepticism. A fundamental problem for inferring evolutionary processes from population genomic variation is that genome variation typically contains very limited information, particularly if we are considering only allele frequencies. The allele frequencies at variable nucleotides most commonly correspond to one or a few copies of a rare allele, with the remainder of the population possessing the second, common allele (with p 1⁄4 1=2N as the most common frequency among variable sites, 2/2N as the next most common, etc.; Nelson et al., 2012; Lek et al., 2016). Consequently, most of the genome will be uninformative about associations of alleles in genotypes and populations (e.g. FST) or between loci (measures of linkage disequilibrium). The maximum possible value of FST is limited by the allele frequency of the major allele (Jakobsson et al., 2013), and much of the variation across the genome will be shaped by global frequencies (and possible information), rather than population structure. Standardization of these statistics that take into account the maximum possible value of FST (Hedrick, 2005; Meirmans & Hedrick, 2011; Whitlock, 2011) or linkage disequilibrium does nothing to affect the information content of the allele frequency variation. Thus, it is possible that covariation of measures of FST and linkage disequilibrium (Hohenlohe et al., 2012) results in part from the information limits set on each by underlying global allele frequencies, rather than reflecting evolutionary processes. Although FST variation among loci can reflect differences in selection and migration that affect differentiation, it is clear that FST poorly tracks variation in these processes. In the future, our measures of recombination will likely be improved as long reads from DNA sequencing will provide direct observations of gametes and meioses in each parent. Given the very limited information in FST, at most loci, identification of regions of the genome that harbour barrier loci will be facilitated by explicit models of gene flow rather than consideration of differentiation (and islands and other metaphors). These models of gene flow will typically need to be complex (incorporating demography, recombination, mutation, possibly complex migration histories, etc.), and analyses will involve comparison of a large number of alternative possible histories. It seems very possible that model choice that today is regarded as comprehensive, complex or both (e.g. Gutenkunst et al., 2009; Christe et al., 2017) will be unacceptably, overly simplistic in the notso-distant future. Beyond the difficulty of inference based on allele frequencies alone, a further challenge to understanding the genetics of barriers between species comes from the complexity of trait genetics. Reproductive isolation between many species is likely to result from many phenotypes that function in different life stages and Correspondence: C. A. Buerkle, Department of Botany, University of Wyoming, 1000 E. University Ave., 82071 Laramie, WY, USA. e-mail: buerkle@uwyo.edu