By Genetic Engineering and Biotechnology News
In this corner: the state-of-the-art enzyme function prediction tool … the Bioinformatic Brawler … the Basic Local Alignment Search Tool for proteins … BLASTp! And in this corner: the contender … the In Silico Kid … contrastive learning–enabled enzyme annotation … CLEAN!
Neither the champ nor the challenger had to be told to have a clean fight. Both had trained to be artificial intelligence (AI) tools that could predict the functions of enzymes based on their amino acid sequences. Both had demonstrated that they could go the distance even if a contest involved unstudied or poorly understood enzymes.
So, what was the outcome? By a decision … CLEAN! It attracted high scores for accuracy, reliability, and sensitivity. CLEAN, according to its creators at the University of Illinois Urbana-Champaign, promises to advance research in genomics, chemistry, industrial materials, medicine, pharmaceuticals, and more.