31 Genotype Concordance between SNP Chip and Imputed Low-Pass Whole-Genome Sequence in Beef Cattle

Abstract Genotypes generated from common single nucleotide polymorphisms (SNP) chips have been the sole source of genomic information for current beef cattle genetic evaluations. More recently, the development of low-pass whole-genome sequencing (LPS) followed by imputation to full sequence and variant calls has provided an alternative option to generate genotypes for inclusion into genetic evaluations. The objective of this study was to estimate the genotype concordance between SNP chip and LPS genotypes. Ear samples, via tissue sampling units, were obtained from 226 individuals. A subset of 109 individuals also had blood samples collected. Samples were sent to GeneSeek (Neogen company, Lincoln, NE) for genotyping (100K or 50K) and LPS and imputation was performed by Gencove. A total of 33,451 and 78,032 SNP overlapping between SNP panel and LPS for ear and blood samples, respectively, were used for further analyses. Genotype concordance was performed by counting the number of matching genotypes per SNP between (1) SNP chip vs. LPS in ear and blood samples, and (2) LPS in blood vs. LPS in ear samples. The average concordances between SNP chip vs. LPS in ear and blood samples were 0.97±0.04 and 0.98±0.06, respectively. The average concordance between LPS in blood vs. ear samples was 0.97±0.03. In conclusion, the concordance was high regardless of the tissue type, and, at least in these data, differences in resulting genetic predictions would not be expected between the use of genotypes derived from SNP chips or LPS. Additional work is warranted to determine sensitivity to relationships between target animals and the reference set used for imputation. The USDA is an equal opportunity provider and employer.