Variability of bovine serum amyloid A3 and somatic cell score

Using a target re-­‐sequencing approach, we iden6fied all SNPs in a region of 20kb of BTA 29 containing the SAA3.2 gene, a candidate gene for mas66s resistance in Italian Holstein cows. A TruSeq Custom Amplicon Assay was designed to resequence 12 kb upstream of the promoter and 1 kb downstream of the 3’-­‐UTR (Miseq NGS technology). The DNA of 95 bulls was extracted, amplified with the custom assay and sequenced. Animals were chosen using a selec6ve genotyping approach according to their soma6c cell score breeding value–SCS(EBV). An individual average coverage threshold of 20X was considered, resul6ng in 52 high and 33 low EBV(SCS) individuals retained. A total of 446 SNPs were iden6fied but only 127 SNPs, with a minor allele frequency (MAF) lower than 0.05, were considered: 92 upstream, 8 in the promoter, 20 in gene and 7 in the downstream region. Associa6on analysis between SNPs and SCS(EBV) was carried out using two different approaches. The first approach was the MAX test proposed for case-­‐control studies by Fridlin et al. (2002), used to verify the associa6on between the binary trait ‘nega6ve-­‐posi6ve tails’ and SNPs. The second approach was the heteroscedas6c effects model (HEM) (Shen et al. 2013). This model was used with the objec6ve to capture gene6c effects that are oren quite small. All analyses were performed in R, using the package bigRR (Shen et al., 2013) and Rassoc (Zang et al., 2010), bootstraping (with 50,000 replicates) to approximate the distribu6on of the MAX test under the null hypothesis of equal genotype distribu6on in the 2 tails as described in Fontanesi et al. (2012) and using a binomial distribu6on. With the MAX test five SNP were found significant using a threshold of p-­‐value <0.1 (0.013-­‐0.075): rs137746604(A/G), rs210417381(C/T) and rs136687125 (C/T), located in the promoter region, rs42175271 (C/G) and rs378094124, located upstream of the promoter. With the HEM method three SNPs iden6fied in MAX test showed a highest heteroscedas6c effect: rs137746604 (0.030), rs136687125(0.012) and rs42175271 (0.017). In addi6on, rs42175273(A/T) and rs384439423(C/T), located upstream of the promoter, had effect > 0.0123. Analysis with different approaches (e.g. Bayesian, GRAMMAR) are underway. Although these results must be confirmed by the analysis of a large number of individuals, this inves6ga6on is our first contribu6on to the iden6fica6on of markers for gene6c resistance to mas66s in Italian Holstein.