Versatile simulations of admixture and accurate local ancestry inference with mixnmatch and ancestryinfer

Abstract

It has become clear that hybridization between species is much more common than previously recognized. As a result, we now know that the genomes of many modern species, including our own, are a patchwork of regions derived from past hybridization events. Increasingly researchers are interested in disentangling which regions of the genome originated from each parental species using local ancestry inference methods. Due to the diverse effects of admixture, this interest is shared across disparate fields, from human genetics to research in ecology and evolutionary biology. However, local ancestry inference methods are sensitive to a range of biological and technical parameters which can impact accuracy. Here we present paired simulation and ancestry inference pipelines, mixnmatch and ancestryinfer, to help researchers plan and execute local ancestry inference studies. mixnmatch can simulate arbitrarily complex demographic histories in the parental and hybrid populations, selection on hybrids, and technical variables such as coverage and contamination. ancestryinfer takes as input sequencing reads from simulated or real individuals, and implements an efficient local ancestry inference pipeline. We perform a series of simulations with mixnmatch to pinpoint factors that influence accuracy in local ancestry inference and highlight useful features of the two pipelines. mixnmatch is a powerful tool for simulations of hybridization while ancestryinfer facilitates local ancestry inference on real or simulated data.

Two new hybrid zones expand the swordtail hybridization model system

“…Pure X. cortezi were collected in January 2020 from the Río Huichihuayán, a fully allopatric population with respect to X. birchmanni (Puente de Huichihuayán, 21°26'9.95"N 98°56'0.00"W, n=42). One previously sequenced pure X. cortezi individual from Las Conchas (21°23'33.30"N 98°59'23.33"W) and seven from el nacimiento de Huichihuayán (21°27'34.10"N 98°58'36.70"W) were included in analyses (Powell et al, 2020;Schumer et al, 2018). Likewise, pure X. birchmanni from Coacuilco (21° 5'51.16"N 98°35'20.10"W) were collected previously for studies of hybridization with X. malinche (Schumer et al, 2018).…”

“…While artificial hybrids, particularly in Drosophila, formed the foundation of early research into the genetic barriers that differentiate species (Coyne & Orr, 1997;Dobzhansky, 1936;Orr & Coyne, 1989), in recent years there has been a renaissance in the study of natural hybrid populations (e.g. Brandvain, Kenney, Flagel, Coop, & Sweigart, 2014;Powell et al, 2020;Sankararaman et al, 2014;Stukenbrock, Christiansen, Hansen, Dutheil, & Schierup, 2012;Turissini & Matute, 2017).…”

The genetic architecture of the sexually selected sword ornament and its evolution in hybrid populations

“…We phenotyped 536 adult male F2 hybrids and collected low-coverage whole-genome sequence data (~0.2X coverage; Materials & Methods). Using a pipeline we previously developed [30], we inferred local ancestry of each individual along the 24 swordtail chromosomes ( Fig S2). Simulations indicated that we expect this approach to have high accuracy given our cross design ( Fig S3; Supporting Information 1).…”

“…To infer local ancestry, we used a pipeline we previously developed called ancestryinfer [30,63]. Briefly, for each individual Illumina reads were mapped to both the X. birchmanni and X. malinche reference genomes; uniquely mapping reads were retained and counts for each allele were tabulated at each ancestry informative site.…”

Model-based genotype and ancestry estimation for potential hybrids with mixed-ploidy

“…Comparisons of parameter estimates from models with different numbers (K) of source populations can guide an understanding of hierarchy and spatial genetic structure and admixture among the sampled individuals. Beyond structure-like models, there is considerable interest in estimates of locus-specific ancestry and introgression, with a corresponding wealth of existing and continuously developing methods in computational statistics (e.g., Sankararaman et al, 2008;Gompert, 2016;Rosenzweig et al, 2016;Ottenburghs et al, 2016;Schumer et al, 2019, for a review, see Gompert et al 2017). These include parametric methods for detecting loci with ancestry that is concordant with the remainder of the genome (e.g., Szymura & Barton, 1986;Gompert & Buerkle, 2011a), or for detecting breakpoints and tracts of ancestry among chromosomal blocks or haplotypes (e.g., Wegmann et al, 2011;Lawson et al, 2012;Sohn et al, 2012;Gompert, 2016).…”