Accuracy of Imputation of Microsatellite Markers from an SNP50K Chip in Spanish Assaf Sheep

Abstract

Background: Traditional and new genotyping technologies must be combined by applying bridge methodologies that avoid double genotyping costs. This study aims to identify and evaluate a reliable approach to precisely impute microsatellite markers from SNP-chip panels to perform parental verifications in sheep. Moreover, we will assess the optimum number of SNPs necessary to accurately impute microsatellite markers to develop a low-density SNP chip for parentage verification in the Assaf sheep breed.Results: A total of 4,423 animals belonging to the Spanish Assaf sheep breed were genotyped for 19 microsatellites and an ovine custom 49,897 SNP array. The accuracy of microsatellite marker imputation, performed with BEAGLE v5.1 software, was assessed with three metrics, namely, genotype concordance (C), genotype dosage (length r2), and allelic dosage (allelic r2), for all imputation scenarios tested (0.5-10 Mb microsatellite flanking SNP windows). The accuracy of our imputation results for the three metrics analyzed for all haplotype lengths tested was higher than 0.90 (C), 0.80 (length r2), and 0.75 (allelic r2). Considering that the objective of the study was to assess a SNP window length that provides the best accuracy for the microsatellite imputation procedure to design an affordable low-density SNP chip for parentage testing, we considered 2 Mb to be the best SNP haplotype length for further analyses (SNPs/window =74.05, C= 0.970; length r2= 0.952, allelic r2=0.899). We additionally evaluated imputation performance under two null models, naive and random, which showed weak genotype concordance averages in comparison with imputed microsatellites (0.41 and 0.15, respectively).Conclusions: We presented for the first time a precise methodology in dairy sheep to impute multiallelic microsatellite genotypes from biallelic SNP markers. The use of a 2 Mb SNP flanking window for each microsatellite has been shown to achieve high accuracy in the imputation procedure while providing a low-density SNP chip that could be cost-effective. The results from this study will undoubtedly have a significant impact on sheep breeders overcoming the problem of parentage verification when different genotyping platforms have been used across generations.

FarmGTEx TWAS-server: an interactive web server for customized TWAS analysis in both human and farm animals

“…Genome-wide association studies (GWAS) have discovered numerous genetic variants associated with complex diseases and traits in both human and livestock populations (1)(2)(3)(4). However, most of these variants are in high linkage disequilibrium (LD) and reside in noncoding regions, which makes it extremely challenging to interpret their underlying molecular mechanisms.…”