Introgression history in sticklebacks and oaks

Speciation theory has many monikers for differential gene flow – migration, introgression, admixture, hybridization, secondary contact. As a homogenizing process, gene flow at large acts to reduce differentiation between populations post-divergence. However, selection and demography affect the rates of gene flow across genomic loci (see my detailed post about differential migration here). Recently, two studies in very different speciation regimes – oaks speciating in sympatry, and threespine sticklebacks in parapatry/allopatry detail differential introgression using large scale reduced representation sequencing.

The impact of selection, gene flow and demographic history on heterogeneous genomic divergence: threespine sticklebacks in divergent environments, Ferchaud and Hansen (2015), Molecular Ecology

Threespine sticklebacks have been described by numerous previous studies for their genomic and morphological adaptations to saline waters, in the presence (and absence) of gene flow from fresh waters. Ferchaud and Hansen (2015) address three phenomena with respect to heterogeneous gene flow and adaptive divergence – (1) genome-wide patterns of differentiation in riverine versus lake populations which freely exchange genes, versus lack gene flow with marine environments respectively, (2) footprints of demographic changes in population sizes (bottlenecks) across genomes, and (3) describing patterns of adaptive divergence in populations showing less drastic (than previously described studies) changes in morphology.

Sampling across a variety of environments, Ferchaud and Hansen use RAD sequencing of 177 sticklebacks and analyze LD, differentiation among populations, population structure, compare models of demographic history, and identify “outlier” loci – loci exhibiting significant differentiation across populations. Ultimately using >23k loci after filtering, their key findings include (a) highest diversity, and lower LD decay levels in marine environments, and lowest diversity and higher LD decay in smaller lakes (without inflow), with intermediate levels of diversity and LD in lakes with riverine inflow (b) high Fst (0.204) among all populations, with varying levels of differentiation among all pairs of populations with similar patterns uncovered by analyses of population structure, (c) stable population size history in marine environments, versus all freshwater populations exhibiting signatures of bottlenecks, all observations pointing towards the role of heterogeneous gene flow in population differentiation and adaptive divergence.

Historical introgression among the American live oaks and the comparative nature of tests for introgression – Eaton et al. (2015), Evolution

The Angel Oak Tree (Quercus virginiana), a species of American Live oak, believed to be over 1500 years old. American live oaks are termed a “worst case scenario for the biological species concept” by Coyne and Orr (2004). Image courtesy – Pinterest

Natural populations of American Live oaks (Quercus series Virentes) hybridize freely, and yet several extant species/subspecies exist across its distribution in a variety of climates and geographies. Eaton et al. (2015) use RADseq data to quantify genetic admixture, and ancestral demography in fourteen American clade oaks using five widely used methods – phylogenetic resolution into clades, inference of population structure under an admixture model, joint inference of tree topology and admixture, D-statistic (ABBA-BABA) tests, and  demographic models by comparison of site frequency spectra. Phylogenetic reconstruction recovered support for three major clades – southeastern US, southwestern clade, and a Central American clade, also obtained using analyses of population structure. Admixture analyses supported admixture edges between Central America, and southeastern US clades, and additional edges between southeastern and southwestern clades. ABBA-BABA tests (non-parametric) showed heterogeneity in patterns of admixture, with significant results restricted to geographically proximal clades. Analyses of SFS differences also reject a hybrid origin for Cuban populations, with indistinguishable support for Central American, and southeastern US origin.

Our analyses demonstrate the difficulty of inferring historical introgression over deep evolutionary time scales…Here, rather than focus on a specific number of individual, we stress the importance of attaining “phylogenetically relevant sampling,” which we define to include several geographic samples from both within species and across species to allow for contrasts that can reveal the presence and geographic extent of admixture.

References:

Ferchaud, AL, and Hansen MM. “The impact of selection, gene flow and demographic history on heterogeneous genomic divergence: threespine sticklebacks in divergent environments.” Molecular Ecology (2015). DOI: http://dx.doi.org/10.1111/mec.13399

Eaton, Deren AR, et al. “Historical introgression among the American live oaks and the comparative nature of tests for introgression.” Evolution (2015). DOI: http://dx.doi.org/10.1111/evo.12758

 

 

 

 

 

 

Share

About Arun Sethuraman

I am a computational biologist, and I build statistical models and tools for population genetics. I am particularly interested in studying the dynamics of structured populations, genetic admixture, and ancestral demography.
This entry was posted in adaptation, evolution, genomics, natural history, phylogenetics, population genetics, selection, speciation, species delimitation and tagged , , , , , . Bookmark the permalink.