Genome Wide Association Study of Beef Traits in Local Alpine Breed Reveals the Diversity of the Pathways Involved and the Role of Time Stratification

Abstract

Knowledge of the genetic architecture of key growth and beef traits in livestock species has greatly improved worldwide thanks to genome-wide association studies (GWAS), which allow to link target phenotypes to Single Nucleotide Polymorphisms (SNPs) across the genome. Local dual-purpose breeds have rarely been the focus of such studies; recently, however, their value as a possible alternative to intensively farmed breeds has become clear, especially for their greater adaptability to environmental change and potential for survival in less productive areas. We performed single-step GWAS and post-GWAS analysis for body weight (BW), average daily gain (ADG), carcass fleshiness (CF) and dressing percentage (DP) in 1,690 individuals of local alpine cattle breed, Rendena. This breed is typical of alpine pastures, with a marked dual-purpose attitude and good genetic diversity. Moreover, we considered two of the target phenotypes (BW and ADG) at different times in the individuals’ life, a potentially important aspect in the study of the traits’ genetic architecture. We identified 8 significant and 47 suggestively associated SNPs, located in 14 autosomal chromosomes (BTA). Among the strongest signals, 3 significant and 16 suggestive SNPs were associated with ADG and were located on BTA10 (50–60 Mb), while the hotspot associated with CF and DP was on BTA18 (55–62 MB). Among the significant SNPs some were mapped within genes, such as SLC12A1, CGNL1, PRTG (ADG), LOC513941 (CF), NLRP2 (CF and DP), CDC155 (DP). Pathway analysis showed great diversity in the biological pathways linked to the different traits; several were associated with neurogenesis and synaptic transmission, but actin-related and transmembrane transport pathways were also represented. Time-stratification highlighted how the genetic architectures of the same traits were markedly different between different ages. The results from our GWAS of beef traits in Rendena led to the detection of a variety of genes both well-known and novel. We argue that our results show that expanding genomic research to local breeds can reveal hitherto undetected genetic architectures in livestock worldwide. This could greatly help efforts to map genomic complexity of the traits of interest and to make appropriate breeding decisions.

Improvement of Genomic Predictions in Small Breeds by Construction of Genomic Relationship Matrix Through Variable Selection

“…It is possible to observe a few clusters in the genomic relationship matrix ( Figure 3 ); however, they are not as straightforward as in SIM. We, therefore, can note that no population structure is present in Rendena breed (i.e., no subpopulations), which is in line with previous research ( Mancin et al, 2022 ). Only 633 eigenvalues explained 98% of G variance; thus, the scenario was closer to SIM2 than SIM1.…”

“…Only 633 eigenvalues explained 98% of G variance; thus, the scenario was closer to SIM2 than SIM1. In addition, we observed an average LD of 0.187 ± 0.107 per chromosome ( Mancin et al, 2022 ).…”

“…Furthermore, Akbarzadeh et al (2021) reported a dramatic decline in performance when the same percentage of SNPs was randomly chosen. We tried preliminary tests of a similar approach—construction of the G matrix using the top 500, 1,000, and 50,000 SNPs ranked by their absolute SNP effect values calculated through back solutions—in Rendena breed; however, we immediately discarded this approach because of the extreme bias and inflated breeding value predictions (these findings are reported by Mancin et al, 2022 in press ). In addition, choosing so few and unrepresentative SNPs reduced a lot the compatibility between A and G matrices, and thus ssGBLUP properties were affected ( Misztal et al, 2013 ).…”

“…The overlap between the two panels was about 60%. Information about data quality control and imputation is reported in greater detail by Mancin et al (2022) . In addition to the previous study, further quality control was performed by removing SNPs with high linkage disequilibrium (>80), using SNPrune ( Calus and Vandenplas, 2018 ): this removed a total of 28,049 SNPs.…”

Genome-wide analysis in the search for candidate genes associated with meat productivity traits in meat-and-dairy goats

“…In a study performed on the Rendena cattle bred in the Italian Alps important for beef production, it was found that SLC12A1, CGNL1, PRTG genes were associated with average daily body weight gain, LOC513941 gene was associated with carcass yield, CDC155 gene was associated with the content of boneless meat in the carcass, NLRP2 gene was associated with both indicators of carcass quality. Metabolic pathways analysis showed that some of the genes were associated with neurogenesis and synaptic signaling in cells, and some were associated with actin synthesis and transmembrane transport [30].…”