A need for standardized reporting of introgression: Insights from studies across eukaryotes

Abstract

With the rise of affordable next-generation sequencing technology, introgression-or the exchange of genetic materials between taxa-has become widely perceived to be a ubiquitous phenomenon in nature. Although this claim is supported by several keystone studies, no thorough assessment of the frequency of introgression across eukaryotes in nature has been performed to date. In this manuscript, we aim to address this knowledge gap by examining patterns of introgression across eukaryotes. We collated a single statistic, Patterson's D, which can be used as a test for introgression across 123 studies to further assess how taxonomic group, divergence time, and sequencing technology influence reports of introgression. Overall, introgression has mostly been measured in plants and vertebrates, with less attention given to the rest of the Eukaryotes. We find that the most frequently used metrics to detect introgression are difficult to compare across studies and even more so across biological systems due to differences in study effort, reporting standards, and methodology. Nonetheless, our analyses reveal several intriguing patterns, including the observation that differences in sequencing technologies may bias values of Patterson's D and that introgression may differ throughout the course of the speciation process. Together, these results suggest the need for a unified approach to quantifying introgression in natural communities and highlight important areas of future research that can be better assessed once this unified approach is met.

Figure 1

Distribution of Patterson's D values sampled across the Eukaryote phylogeny of phyla (Hedges et al. 2015), with only phyla with labeled data. The number of records per taxonomic class (n) and number of source studies (s) are listed by the taxon labels, while the distribution of Patterson's D values is displayed on the inset. Significant values of Patterson's D (determined either by P‐value, Z score, or significance stated in source paper) are colored black, while nonsignificant values are colored red.

Figure 2

The relationship between genetic distance and A) Significance of introgression tests and B) Magnitude of Patterson's D. Both relationships are significant in linear mixed models, but phylogenetic bootstrap estimates of effects overlap 0 for significance of Patterson's D (Figure S6) while remaining significant for the magnitude of Patterson's D (Figure S9). Solid lines represent the best fit from mixed models, while dashed lines show naïve linear model fits – accounting for the random effects of species pairs and reference reverses the slope in both cases.

Figure 3

Phylogenetically independent clusters of introgression reports versus genetic distance. Reducing the data to 100 clusters of introgression events, we find no significant relationship between genetic distance and either significance or magnitude of Patterson's D. Dashed lines show linear model fits (inclusion of reference as a random effect was precluded by some clusters consisting of reports from many references).

Figure 4

Sequencing type and Patterson's D. Studies using reduced representation sequencing (GBS/RAD) report significantly larger values of Patterson's D than studies that use whole genome sequences.