Tissue-specific RNA expression data enhances genomic prediction of methane emissions in sheep

<p dir="ltr">Reducing methane emissions in sheep through selective breeding requires accurate genomic prediction methods. This study investigated whether incorporating RNA tissue expression data could enhance prediction accuracy for methane emissions. Analysis was conducted on 1,366 sheep, including RNA sequencing data from three tissues (longissimus muscle, rumen wall, and rumen papillae) in a subset of 48 animals. To address the high dimensionality of RNA expression data (>20,000 genes), we employed principal component analysis (PCA) and selected components based on their correlation with methane emissions. These selected components were integrated as intermediate traits in a Neural Network Genomic Best Linear Unbiased Prediction (NNGBLUP) model, evaluating up to nine principal components (PCs) sequentially for each tissue. While the traditional GBLUP approach demonstrated robust performance within our flock sheep population, the integration of tissue-specific RNA expression data through selected PCs provided modest improvements in prediction accuracy. These findings suggest that RNA tissue expression data can complement existing genomic prediction methods, despite constraints from limited sample size and genetic diversity. Our results highlight the potential of incorporating molecular phenotypes in breeding programs aimed at reducing methane emissions in sheep.</p>