Despite the identification of candidate genes influencing milk protein, the connections between genes and regulatory path- ways remains elusive. This study aimed integrate findings from genome-wide association studies (GWAS) and RNA sequencing (RNA-Seq) through meta-analysis to pinpoint single nucleotide polymorphisms (SNPs) and genes responsible for high and low protein yield in cows. Previous GWAS and RNA-Seq analyses had identified 663 SNPs and 1106 genes (P < 0.05). Twenty SNPs from GWAS, 10 genes from RNA-Seq, and 49 SNP/gene associations from both datasets, were identified using meta-analysis. Meta-analysis validated several SNPs previously identified through GWAS, such as rs135549651 (P = 2.6 × 10−256 ), rs109146371 (P = 3.1 × 10−208 ), rs109350371 (P = 4.0 × 10−207 ), and rs109774038 (P = 8.6 × 10−587 ). Genes identified in RNA-Seq exper- iments, including NR4A1 (P = 3.2 × 10−7 ), ATF3 (P = 9.6 × 10−7 ), CDH16 (P = 9.9 × 10−7 ), VEGFA (P = 1.0 × 10−6 ), and SAA3 (P = 7.3 × 10−11 ), were confirmed. The combined GWAS and RNA-Seq datasets highlighted CCND2 (P = 8.9 × 10−111 ), MAPK15 (P = 1.3 × 10−151 ), and CPSF1 (P = 1.2 × 10−306 ) as the most significant genes. Additionally, significant gene ontology (GO) terms, including ionizing radiation (P = 1.5 × 10−4 ), nuclear pore cytoplasmic filaments (P = 9.4 × 10−5 ), and phenylalanine 4-monooxygenase activity (P = 1.4 × 10−5 ), were identified. In conclusion, the integration of GWAS and RNA-Seq, coupled with GO enrichment, allowed identification of candidate SNPs and genes with higher accuracy. These findings improve our knowledge about genomic architecture of milk protein and enhance evaluation of Holstein cows.

Keywords