Candidate gene approaches provide tools for exploring and localizing causative genes affecting quantitative traits and the underlying variation may be better understood by determining the relative magnitudes of effects of their polymorphisms. Diacyglycerol O-acyltransferase 1 (DGAT1), fatty acid binding protein (heart) 3 (FABP3), growth hormone 1 (GH1), leptin (LEP) and thyroglobulin (TG) have been previously identified as genes contributing to genetic control of subcutaneous fat thickness (SFT) in beef cattle. In the present research, Bayesian model selection was used to evaluate effects of these five candidate genes by comparing competing non-nested models and treating candidate gene effects as either random or fixed. The analyses were implemented in SAS to simplify the programming and computation. Phenotypic data were gathered from a F2 population of Wagyu x Limousin cattle. The five candidate genes had significant but varied effects on SFT in this population. Bayesian model selection identified the DGAT1 model as the one with the greatest model probability, whether candidate gene effects were considered random or fixed, and DGAT1 had the greatest additive effect on SFT. The SAS codes developed in the study are freely available and can be downloaded at: http://www.ansci.wsu.edu/programs/....

Evaluation of Candidate Gene Effects for Beef Backfat via Bayesian Model Selection

10.1007/s10709-005-5255-1