Genome Wide Association Studies with Different Weighting Approaches Reveals Genomic Windows Associated with Meat Quality Traits in Beef Cattle.

Abstract

<b>Background/Objectives:</b> Genome-wide association studies (GWAS) based on single-step genomic BLUP (ssGBLUP) commonly assume equal single nucleotide polymorphism (SNP) variances, which may not reflect the biological architecture of complex traits. Alternative weighting strategies can increase detection power but may affect stability. This study evaluated how different SNP weighting approaches influence genomic region detection and biological interpretation of ribeye area (REA) and subcutaneous fat thickness (SFT) in Guzerá cattle. <b>Methods:</b> Phenotypic records from 2729 animals and genotypes from 1405 individuals (43,039 SNPs after quality control) were analyzed. Heritabilities were estimated using Restricted Maximum Likelihood (REML), and GWAS were conducted under five approaches: unweighted method (UM), quadratic method (QM), and three Non-Linear A strategies with weighting constants (1.125, 1.2, and 1.5). Genomic windows of 20 adjacent SNPs explaining ≥0.5% of the additive genetic variance (AGV) were considered significant. Recurrent regions were prioritized, and functional enrichment analyses (KEGG, GO, and MeSH) were performed. <b>Results:</b> Heritability estimates were moderate for REA (0.26 ± 0.05) and SFT (0.22 ± 0.04). Weighted approaches increased detection sensitivity. For REA, UM identified 10 windows, whereas QM and A_1.5 detected 24 and 31 windows. For SFT, UM identified 8 windows, while QM and A_1.5 detected 30 and 23 windows. Recurrent chromosomes included 2, 4, 6, 12, 16, 19, and 22 for REA, and 2, 3, 5, 7, 11, 17, and 22 for SFT. Key genes included <i>AKT3</i>, <i>NOS2</i>, and <i>MSTN</i>. Enrichment highlighted pathways related to muscle growth and lipid metabolism. <b>Conclusions:</b> SNP-weighted GWAS increased detection sensitivity but involved trade-offs between signal amplification and stability. Integrating weighting strategies improves biological interpretation and supports robust candidate gene identification for genomic selection.