Applying the algorithm for Proven and young in GWAS Reveals high polygenicity for key traits in Nellore cattle

Abstract

Background: Identifying genomic regions associated with traits of interest and their biological processes provides valuable insights into the phenotypic variability of these traits. This study aimed to identify candidate genes and genomic regions associated with 16 traits currently evaluated by the Brazilian Association of Zebu Breeders (ABCZ). These traits include reproductive traits such as age at first calving (AFC), stayability (STAY), and scrotal circumference at 365 (SC365) and 450 days (SC450). Growth traits include birthweight (BW), expected progeny difference for weight at 120days of age (EPD120), as well as weight at 120 (W120), 210 (W210), 365 (W365), and 450 days of age (W450). Carcass traits include body conformation (BC), finishing score (FS), marbling (MARB), muscularity (MUSC), finishing precocity (FP), and ribeye area (REA).

Methods: A dataset containing 304,782 Nellore cattle genotyped with 437,650 SNPs (after quality control) was used for this study. The Algorithm for Proven and Young (APY), implemented in the PREGSF90 software, was used to compute the $G_{APY}^{-1}$ matrix using 36,000 core animals (which explained 98% of the variance in the genomic matrix). Subsequently, the SNP solutions were estimated by back-solving the Genomic Estimated Breeding Values (GEBVs) predicted by ABCZ using the single-step GBLUP method. Genomic regions were identified using sliding windows of 175 consecutive SNPs, and the top 1% genomic windows, ranked based on their proportion of the additive genetic variance, were used to annotate positional candidate genes and genomic regions associated with each of the 16 traits.

Results: The top 1% windows for all traits explained between 2.779% (STAY) to 3.158% (FP) of the additive genetic variance, highlighting the polygenic nature of these traits. Functional analysis of the candidate genes and genomic regions provided valuable insights into the genetic architecture underlying these traits in Nellore cattle. For instance, our results revealed genes with important functions for each trait, such as SERPINA14 (plays a key role for the endometrial epithelium) identified for AFC, HSPG2 (associated with morphological development and tissue differentiation) identified for BW, among others.

Conclusion: We identified genomic regions and candidate genes, some of which have been previously reported in the literature, while others are novel discoveries that warrant further investigation. These findings contribute to gene prioritization efforts, facilitating the identification of functional candidate genes that can enhance genomic selection strategies for economically important traits in Nellore cattle.

Keywords: GWAS; candidate genes; genomic estimated breeding value; genomic regions; single step.

FIGURE 1

(a–p) Manhattan plot showing the top 1% windows with the highest proportion of genetic variance explained across chromosomes. The red dot represents the top 1% region associated with trait of interest. The Y axis represents “% of variance” explained for each chromosome.