Approaches for validating genomic relationships and mapping of deletions among smallholder dairy cattle populations using SNP genotypes.

Sanarana, Y. P.*¹, Maiwashe, A.¹, Scholtz, M. M.¹, Banga, C. B.²

¹ Agricultural Research Council-Animal Production, Pretoria, Irene, 0062, South Africa
² Botswana University of Agriculture and Natural Resources, Department of Animal Sciences, 0027, Gaborone, Botswana

Genealogical information is an essential tool for carrying out many applications for precise genetic improvement programs. Nevertheless, the lack of accurate pedigree records under uncontrolled practices of smallholder farming systems makes genomic selection unfeasible. The use of molecular technologies, particularly single nucleotide polymorphism (SNP) genotypes, provide prolific source of information which could assist to capture hidden genetic relationships and in turn be used to map genomic deletions responsible for substantial economic gains or losses in the dairy industry. This study was undertaken to validate genomic relationships and utilize correct genealogy records to map deletions among the genotyped smallholder dairy animals. Genomic data comprising of 1, 300 animals from five herds with 60K SNP variants from the Affymetrix bovine microarray was subjected to quality control measures. Only samples with a call rate more than 90% and ~48 000 autosomal SNPs with a call rate ≥ 95%, minor allele frequency (MAF) ≥ 0.05, and P-value for the Hardy-Weinberg equilibrium test ≥10−6 were used for further analysis. Each animal genotype was compared with all other genotypes within a herd to validate genealogical records and to detect unidentified true genomic relationships using R software and awk run script. If SNP conflicts were concentrated on a single chromosome, a chromosomal deletion was captured. Comparison of each genotype with all the others using thousands of markers consume computer space. Thus, a minimum of 300 SNPs was selected for parentage to verify and discover genetic connections within the populations. Pedigree errors ranging from 15% to 45% were detected and a total of 82 genetic relationships actually never recorded were discovered suggesting some recording issues and switching of calves at birth on the farms. The use of 300 set of SNPs indicated that excluding cases of three opposite homozygotes (>1%) could eliminate 99.7% of genotypes without eliminating any confirmed parent-progeny relationships. A number of 13 283 Mendelian mismatches between parent-progeny pairs associated with genomic deletions were observed across 29 chromosomes. Ten SNPs with highest frequency of error occurrences (2000 to 8500) were detected from seven chromosomes exhibiting the regions associated with production, reproduction and the loss-to-function variants. These results indicate that the genealogy recovery and mapping of deletions from the genomic data are essential tools to increase genetic improvement on smallholder herds which in turn could lead to an achievable genomic selection.

Keywords: SNP, polymorphism, genetic improvement