Abstract:

Pork quality is crucial for consumer health and the sustainable development of the swine industry. Traditional breeding methods face limitations in improving meat quality due to insufficient genetic understanding of these traits. With the development of high-throughput technologies, genomic selection(GS) technology has emerged as a key tool for breaking through the genetic improvement bottleneck in meat quality, owing to its high precision and low cost. This paper systematically reviews the development of GS models, categorizing various types of models and discussing their predictive accuracy. Several challenges remain: the technology for precise in vivo phenotypic data collection has yet to be fully realized; existing models show insufficient adaptability to cross-population and cross-environment data; and the integration of multi-omics mechanisms and epigenetic networks into breeding systems is still lacking. Future research should focus on developing dynamic phenotypic sensing technologies, constructing interpretable deep learning frameworks, and creating multi-omics-based predictive models. By establishing a genetic evaluation system covering the entire breeding-production-processing industry chain, this study aims to provide a theoretical foundation and technical pathway for the innovation of high-quality pig germplasm in China, enabling cross-scale analysis from genomics to phenomics.

Key words: pig, meat quality, genomic selection, models comparison, genetic improvement