Improvements in living standards have resulted in consumers having higher expectations for chicken meat quality. This is particularly true in Asia, where there is high consumer preference for local breeds. Nothing is presently known about the effectiveness of using genomic selection (GS) strategies in chickens to genetically improve meat quality traits that cannot be measured in living potential parents. In this study, 724 Beijing-You chickens were used as a training population; all were genotyped using Illumina 60K SNP chips, and intramuscular fat content in breast muscle (IMF br ) was measured. Birds in the GS line were selected based on genomic estimated breeding values, IMF br being the sole trait. Genetic progress in one generation was compared to that from conventional family-based selection, and both were evaluated against random-bred controls. Results showed that relative to the random-bred controls, IMF percentage was improved 9.62% using GS, comparable to the 10.38% improvement using family-based selection. We quantified the effectiveness of GS when applied to a meat quality trait with low heritability in chickens. We plan to introduce custom SNP chips, appropriate for native chicken breeds in China, to assist in applying GS in local breeding and accelerate genetic gain.

Genomic prediction, which is based on solving linear mixed-model (LMM) equations, is the most popular method for predicting breeding values or phenotypic performance for economic traits in livestock. With the need to further improve the performance of genomic prediction, nonlinear methods have been considered as an alternative and promising approach. The excellent ability to How to cite this article: Xiang T, Li T, Li J, Li X, Wang J. Using machine learning to realize genetic site screening and genomic prediction of productive traits in pigs.