Therefore, we used 5′ and 3′ flanking regions of 1 Mb around a significant SNP or up- and downstream from a significant haplotype block to search for candidate genes, which could be responsible for the observed significant associations with SCS.
Notably, all significant (global p<0.05) haplotype blocks that withstood permutation tests in UNPHASED for Dictator Giving and the SVO were also significantly associated using the PBAT multivariate statistic which considers the DG and SVO simultaneously (p<0.05) (see FIGURE 5, FBAT-GEE column) strongly confirming the results observed in UNPHASED.
Then, a Bonferroni test was performed to test phenotypic differences between either genotype or haplotype groups of significant SNPs or haplotype blocks, respectively, using SAS® software, version 9.2 (SAS® Institute Inc., 2008, Cary, NC, USA).
Linkage analysis of these significant SNPs revealed three haplotype blocks of 495 kb, 376 kb and 492 kb, respectively, in the 11.93 Mb region.
Fifteen annotated genes were near or contained the significant SNPs of these three haplotype blocks.
Three of the eight haplotype blocks showed significant association with disease (global P = 0.0067, 0.026 and 0.0386).
LD analysis showed that these multiple significant SNPs were located in two haplotype blocks that spanned 381 and 124 kb, respectively, of which the most significant SNP for FI2, i.e. rs315069556, was in block 2 (Fig. 2).
The significant difference in haplotype block length suggests that selection on this region has been intensive in the PD and BL breeds.
The haplotype blocks in the significant region were constructed using Haploview [ 22] to examine the LD in the region and among significant DArT-seq markers.
