Genome Wide Association Study of Somatic Cell Score in Holstein and Holstein x Jersey crossbred cows from Argentina

NANI JP; AMADIO AF; RASCHIA MA; CARIGNANO HA; POLI MA; CALVINHO LF

Guarujá, Sao Paulo

Simposio; 5th International Symposium on Functional Genomics; 2013

Introduction: Mastitis is the most common disease of dairy cattle,causing high economic loses. Somatic cell count (SCC) is an indirect method of diagnosis, positively correlated with the inflammatory response of the mammary gland and closely associated with udder health. For monitoring udder health and performing genetic evaluations the SCC is commonly log-transformed to somatic cell score (SCS). Genome-Wide Association Study (GWAS) is an efficient way to evaluate genetic variants throughout the genome in order to identify positional candidate genes for traits. Objetive: The aim of this work was to evaluate different parameters calculated from the lactation SCS in order to identify associated SNPs in a GWAS. Methods: Phenotypic data consisted of 3,530 SCC records from 534 Holstein and Holstein X Jersey crosses, daughters from 22 bulls from 14 commercial dairy farms located in the Central Dairy Area of Argentina. Genotyping was carried out using BovineSNP50v2 (Illumina). Quality control was performed on the genotypic data. After filtering by call rate (individuals and SNPs), allele frequency and Hardy-Weinberg equilibrium, genotypic data consisted of 39,255 SNPs. Using trios 2563 mendelian errors were detected and set to missing genotypes in order to impute them with BEAGLE software. Four different variables were calculated from the SCS per lactation: the arithmetic mean of lactation controls (AM), the geometric mean (GM), the maximum value (MAX) and the arithmetic mean of the top 3 values (TOP3). The 4 variables were evaluated with two different animal models using WOMBAT program. Associations between each individual SNP and the corrected phenotype were estimated using PLINK prog ram. To account for the risk of false positives, a false discovery rate threshold of 0.20 was set. Results: The analyses showed that AM and GM have a strong correlation between them (0.90) and also TOP3 and MAX (0.83). AM and GM were significantly associated (p