Genetics

Genetics Language models

Genomic language model predicts protein co-regulation and function

The authors show in their paper the ability to train genomic language models on top of protein language modesl (ESM2) "on millions of metagenomic scaffolds to learn the laten functional and regulatory relationships between genes." Their reveal "a promising approach to encode functional semantics and regulatory syntax of genes in ther genomic context and uncover complex relationships between genes in a genomic region."