Evaluation of a Chicken 600K SNP genotyping array in non-model species of grouse

Piotr Minias,Jeff A Johnson,Linda A Whittingham,Peter O Dunn,Sara J Oyler-Mccance

doi:10.1038/s41598-019-42885-5

Abstract

The use of single nucleotide polymorphism (SNP) arrays to generate large SNP datasets for comparison purposes have recently become an attractive alternative to other genotyping methods. Although most SNP arrays were originally developed for domestic organisms, they can be effectively applied to wild relatives to obtain large panels of SNPs. In this study, we tested the cross-species application of the Affymetrix 600K Chicken SNP array in five species of North American prairie grouse (Centrocercus and Tympanuchus genera). Two individuals were genotyped per species for a total of ten samples. A high proportion (91%) of the total 580 961 SNPs were genotyped in at least one individual (73–76% SNPs genotyped per species). Principal component analysis with autosomal SNPs separated the two genera, but failed to clearly distinguish species within genera. Gene ontology analysis identified a set of genes related to morphogenesis and development (including genes involved in feather development), which may be primarily responsible for large phenotypic differences between Centrocercus and Tympanuchus grouse. Our study provided evidence for successful cross-species application of the chicken SNP array in grouse which diverged ca. 37 mya from the chicken lineage. As far as we are aware, this is the first reported application of a SNP array in non-passerine birds, and it demonstrates the feasibility of using commercial SNP arrays in research on non-model bird species.

Highlights

DNA genotyping arrays with hundreds of thousands single nucleotide polymorphisms (SNPs) are commonly used in genetic studies of model species, but relatively few arrays have been developed in other species because of their high costs of development
Cross-species application of the domestic dog CanineHD BeadChip to seals amplified 19% of 173 K loci, but fewer than 200 loci were polymorphic[13]. Another issue with cross-species use of SNP arrays is ascertainment bias, in which the sample of SNPs does not represent the overall pattern of genetic diversity, in this case because the SNP array was developed in a different species. This difference is likely to lead to an over-representation of polymorphisms shared between species or polymorphisms resulting from artificial selection, when SNPs originally developed for domestic species are cross-amplified in wildlife
Ascertainment bias might lead to spatial clustering of polymorphisms around selectively non-neutral SNPs, but most commercial SNP arrays developed for domestic animals are based on random selections of evenly spaced SNPs which may help to lessen this bias[13]

Summary

Introduction

DNA genotyping arrays with hundreds of thousands (or more) single nucleotide polymorphisms (SNPs) are commonly used in genetic studies of model species, but relatively few arrays have been developed in other species because of their high costs of development. Cross-species application of the domestic dog CanineHD BeadChip to seals (divergence time ~44 mya) amplified 19% of 173 K loci, but fewer than 200 loci were polymorphic[13]. Another issue with cross-species use of SNP arrays (and other markers) is ascertainment bias, in which the sample of SNPs does not represent the overall pattern of genetic diversity, in this case because the SNP array was developed in a different species. Our cross-species application of chicken SNPs took advantage of the expected high synteny between the genomes of chicken and grouse (see Methods for details) and, to the best of our knowledge, this is the first study to test the Affymetrix 600K SNP array on a species other than Gallus gallus

Methods

Results

Conclusion