Enzyme function prediction using contrastive learning

Enzyme function annotation is a fundamental challenge, and numerous computational tools have been developed. However, most of these tools cannot accurately predict functional annotations, such as enzyme commission (EC) number, for less-studied proteins or those with previously uncharacterized functions or multiple activities. We present a machine learning algorithm named CLEAN (contrastive learning-enabled enzyme annotation) to assign EC numbers to enzymes with better accuracy, reliability, and sensitivity compared with the state-of-the-art tool BLASTp. The contrastive learning framework empowers CLEAN to confidently (i) annotate understudied enzymes, (ii) correct mislabeled enzymes, and (iii) identify promiscuous enzymes with two or more EC numbers-functions that we demonstrate by systematic in silico and in vitro experiments. We anticipate that this tool will be widely used for predicting the functions of uncharacterized enzymes, thereby advancing many fields, such as genomics, synthetic biology, and biocatalysis.

Automated summary

CLEAN is a machine learning algorithm based on contrastive learning that can accurately predict functional annotations of enzymes, especially for less-studied or uncharacterized proteins with multiple activities. It can confidently annotate understudied enzymes, correct mislabeled enzymes, and identify promiscuous enzymes with multiple EC numbers. This tool is of great significance in predicting the functions of uncharacterized enzymes.