triangulaR: an R package for identifying AIMs and building triangle plots using SNP data from hybrid zones.

Hybridization provides a window into the speciation process and reshuffles parental alleles to produce novel recombinant genotypes. Presence or absence of specific hybrid classes across a hybrid zone can provide support for various modes of reproductive isolation. Early generation hybrid classes can be distinguished by their combination of hybrid index and interclass heterozygosity, which can be estimated with molecular data. Hybrid index and interclass heterozygosity are routinely calculated for studies of hybrid zones, but available resources for next-generation sequencing datasets are computationally demanding and tools for visualizing triangle plots are lacking. Here, we provide a resource for identifying ancestry-informative markers (AIMs) from single nucleotide polymorphism (SNP) datasets, calculating hybrid index and interclass heterozygosity, and visualizing the relationship as a triangle plot. Our methods are implemented in the R package triangulaR. We validate our methods on an empirical dataset and simulations of genetic data from a hybrid zone between two parental groups at low, medium, and high levels of divergence. triangulaR provides accurate and precise estimates of hybrid index and interclass heterozygosity with sample sizes as low as five individuals per parental group, and similar levels of error as another program for hybrid index and interclass heterozygosity estimation, bgchm. We explore various allele frequency difference thresholds for AIM identification, and how this threshold influences the accuracy and precision of hybrid index and interclass heterozygosity estimates. We contextualize interpretation of triangle plots by describing theoretical expectations under Hardy-Weinberg Equilibrium and provide recommendations for best practices for identifying AIMs and building triangle plots.