Genome-wide detection and genomic selection based on genotype-by-environment interaction-associated signals for yak...

Through long-term natural and artificial selection, yaks (Bos grunniens) have evolved diverse coat colors with important aesthetic and economic value. Previous studies indicate that these traits may also serve as important indicators of adaptation to cold, high-altitude environments. However, the genetic mechanisms underlying these traits remain unclear.

Here, we systematically analyzed the genetic architecture of coat color in 511 yaks by combining red-green-blue (RGB)-based quantitative phenotyping with the QingXin 1st 30 K single nucleotide polymorphism (SNP) genotyping chip. RGB components were highly correlated within body regions, but moderately correlated between the head and rump, indicating region-specific pigmentation patterns. Genome-wide association studies (GWAS) of head and rump coat color using five models (general linear model (GLM), mixed linear model (MLM), multiple loci mixed model (MLMM), fixed and random model circulating probability unification (FarmCPU), and Bayesian information and linkage disequilibrium (LD) iteratively nested keyway (BLINK)), together with genotype-by-environment interaction (GbyE) analyses revealed region-specific genetic regulation. Candidate genes associated with yak coat color included KIT and DISP3. KIT was detected in both GWAS and GbyE analyses, underscoring its potential key role in coat color. Heritability (h²) was higher for the head coat color (42.8%) than for the rump (18.0%). In addition, a high-LD block was detected near KIT, further supporting its role as a major candidate gene. Genomic selection (GS) using gBLUP and Bayesian generalized linear regression (BGLR) showed moderate-to-high accuracy, with higher accuracy under random than settings than aligned settings. Genomic-estimated breeding values (gEBVs) were strongly correlated between the head and rump regions, indicating consistent genetic regulation across body regions.

Our study demonstrated that RGB-based phenotyping combined with GWAS and GbyE analyses could identify loci associated with coat color and potential GbyE interactions. GS results indicate that coat color traits are heritable and may inform precise breeding strategies.