Recent introgression between Taiga Bean Goose and Tundra Bean Goose results in a largely homogeneous landscape of genetic differentiation

Abstract

Several studies have uncovered a highly heterogeneous landscape of genetic differentiation across the genomes of closely related species. Specifically, genetic differentiation is often concentrated in particular genomic regions ("islands of differentiation") that might contain barrier loci contributing to reproductive isolation, whereas the rest of the genome is homogenized by introgression. Alternatively, linked selection can produce differentiation islands in allopatry without introgression. We explored the influence of introgression on the landscape of genetic differentiation in two hybridizing goose taxa: the Taiga Bean Goose (Anser fabalis) and the Tundra Bean Goose (A. serrirostris). We re-sequenced the whole genomes of 18 individuals (9 of each taxon) and, using a combination of population genomic summary statistics and demographic modeling, we reconstructed the evolutionary history of these birds. Next, we quantified the impact of introgression on the build-up and maintenance of genetic differentiation. We found evidence for a scenario of allopatric divergence (about 2.5 million years ago) followed by recent secondary contact (about 60,000 years ago). Subsequent introgression events led to high levels of gene flow, mainly from the Tundra Bean Goose into the Taiga Bean Goose. This scenario resulted in a largely undifferentiated genomic landscape (genome-wide F_ST = 0.033) with a few notable differentiation peaks that were scattered across chromosomes. The summary statistics indicated that some peaks might contain barrier loci while others arose in allopatry through linked selection. Finally, based on the low genetic differentiation, considerable morphological variation and incomplete reproductive isolation, we argue that the Taiga and the Tundra Bean Goose should be treated as subspecies.

Fig. 1. Population genetic structure of Taiga and Tundra Bean Goose.

a Map of sampling locations of migrating Taiga and Tundra Bean Goose. b Principal component analysis (PCA) and c ADMIXTURE-analyses show clear genetic differentiation between Taiga and Tundra Bean Goose. Drawings used with permission of the Handbook of the Birds of the World (del Hoyo et al. 2018).

Fig. 2. Distribution of summary statistics.

a Relative divergence (F_ST), b absolute divergence (d_XY), c nucleotide diversity (π) and d Tajima’s D. Correlations between e F_ST and d_XY and between f F_ST and nucleotide diversity.

Fig. 3. The genomic landscape of Taiga and Tundra Bean Goose, with sequences aligned to chicken chromosomes, for relative divergence (F_ST), absolute divergence (d_XY) and nucleotide diversity (π).

The colors in the nucleotide diversity tracks correspond to the Taiga (blue) and the Tundra Bean Goose (red). Only the first three chromosomes and the Z-chromosome are depicted, a complete picture is provided in Supplementary Fig. S3. The highlighted areas (gray boxes) show examples of differentiated islands.

Fig. 4. Demographic analyses point to recent asymmetrical gene flow.

a Four different demographic scenarios and their likelihood scores that were tested with DADI. The most likely model concerns isolation with recent asymmetrical gene flow (log-likelihood=−31,804). b Comparison between the actual data and the simulated model and c the parameters (in 2N-units) estimated in the most likely demographic model.

Fig. 5. Decision tree for classification of species and subspecies based on reproductive isolation, genetic differentiation and morphology.

The black arrows indicate the route followed to determine the taxonomic position of the Taiga and the Tundra Bean Goose.