Increased genetic marker density reveals high levels of admixture between red deer and introduced Japanese sika in Kintyre, Scotland.

S Eryn Mcfarlane,Darren C Hunter,Rebecca Holland,Josephine M Pemberton,Helen V Senn,Stephanie L Smith,Jisca Huisman

doi:10.1111/eva.12880

Abstract

Hybridization is a natural process at species range boundaries, but increasing numbers of species are hybridizing due to direct or indirect human activities. In such cases of anthropogenic hybridization, subsequent introgression can threaten the survival of native species. To date, many such systems have been studied with too few genetic markers to assess the level of threat resulting from advanced backcrossing. Here, we use 44,999 single nucleotide polymorphisms (SNPs) and the ADMIXTURE program to study two areas of Scotland where a panel of 22 diagnostic microsatellites previously identified introgression between native red deer (Cervus elaphus) and introduced Japanese sika (Cervus nippon). In Kintyre, we reclassify 26% of deer from the pure species categories to the hybrid category whereas in the NW Highlands we only reclassify 2%. As expected, the reclassified individuals are mostly advanced backcrosses. We also investigate the ability of marker panels selected on different posterior allele frequency criteria to find hybrids assigned by the full marker set and show that in our data, ancestry informative markers (i.e. those that are highly differentiated between the species, but not fixed) are better than diagnostic markers (those markers that are fixed between the species) because they are more evenly distributed in the genome. Diagnostic loci are concentrated on the X chromosome to the detriment of autosomal coverage.

Highlights

Hybridization is of rising concern to conservation biologists (Allendorf, Leary, Spruell, & Wenburg, 2001; Grabenstein & Taylor, 2018; Rhymer & Simberloff, 1996)
The limited detection power of the marker panels that are typically used in studies of anthropogenic hybridization is insufficient to de‐ tect ongoing introgression and it seems likely that many thousands of markers will be necessary to accurately identify the majority of hybrid individuals (McFarlane & Pemberton, 2019)
When we excluded markers based on linkage disequilibrium (LD) (r2 < .3, 12K single nucleotide polymorphisms (SNPs) remain‐ ing), we found a very high correlation between admixture scores with and without this LD pruning and that only 6 individuals were assigned to a difference species in the LD pruned analysis

Summary

| INTRODUCTION

Hybridization is of rising concern to conservation biologists (Allendorf, Leary, Spruell, & Wenburg, 2001; Grabenstein & Taylor, 2018; Rhymer & Simberloff, 1996). With each generation of backcrossing, the proportion of individu‐ als that are homozygous at diagnostic markers for alleles from one parental species increases, with the power to detect backcrosses sharply decreasing with each additional generation (Boecklen & Howard, 1997) For these reasons, it has been suggested that when determining how many markers to genotype researchers consider divergence between hybridizing species (as suggested by (Vähä & Primmer, 2006)), and the number of elapsed genera‐ tions since hybridization has begun and the recombination rate of the taxa studied (McFarlane & Pemberton, 2019). Anthropogenic hybridization between red deer and Japanese sika provides an opportunity to use SNP markers to search for cryptic backcrossed individuals that were previously identified as parental individuals. Longitude of different classes of marker (diagnostic, ancestry informative and random) in determining hybrid status and understand the difference arising from marker number and genomic location

| METHODS

Findings

| DISCUSSION