Bayesian hybrid index and genomic cline estimation with the R package gghybrid

Abstract

Admixture, the creation of individuals with combined genomic material from multiple differentiated source populations, is now known to be a dominant evolutionary force. Admixture increases polymorphism and can generate novel phenotypes and selection pressures, often leading to both novel adaptation and reproductively isolated hybrid taxa. When a large variety of recombinant types and admixture proportions between two source populations exist, both geographic and genomic cline analysis are suitable methods for inferring biased, restricted, or excessive gene flow at individual loci into the foreign genomic background. Hence, cline analysis can provide evidence for reproductive isolation, selection across an environmental transition, balancing selection, and adaptive introgression, in natural hybridizing populations. Of the two cline methods, genomic cline analysis has fewer assumptions and is suitable in a wider variety of circumstances. Here, I introduce gghybrid, an R package for Bayesian estimation of genome-wide hybrid index and locus-specific genomic clines using bi-allelic data, suitable for both small and large datasets. gghybrid uses Buerkle’s likelihood formula to estimate hybrid index and Fitzpatrick’s logit-logistic genomic cline function to infer restricted, extreme, or biased gene flow. It employs the commonly available Structure file format for data input, is highly parallelizable, and allows use of admixture proportions estimated from other software. Parameters can be pooled across test subjects, or their values fixed, and model comparison carried out using both AIC and waic. Here, I describe the functions, pipeline, and statistical properties of gghybrid.