Learning functional conservation between pig and human to decipher evolutionary mechanisms underlying gene expression and complex trait

Abstract

The assessment of genomic conservation between human and pig at the functional level can help understand and improve the potential of pig as a human biomedical model. To address this, we developed a Deep learning-based approach to learn the Genomic Conservation at the Functional level (DeepGCF) between species by integrating 386 and 374 epigenome and transcriptome profiles from human and pig, respectively. DeepGCF demonstrated a better prediction performance compared to the previous functional conservation prediction method. In addition, we showed that the resulting DeepGCF score captures the functional conservation by examining DeepGCF on chromatin states, sequence ontologies, and regulatory variants. Regions with higher DeepGCF score play a more important role in regulatory activities and show heritability enrichment in human complex traits and diseases. Our DeepGCF approach shows a promising application on the comparison of cross-species functional conservation, and the model framework can be easily adapted to other species. By expanding the model to integrate the functional profiles of multiple species, including human, mouse, pig, cattle, and other livestock animals in the future, the functional conservation information will provide additional insight into the genetic and evolutionary mechanisms behind complex traits and diseases.