Enzymes enter automated synthesis-planning picture

Abstract

IBM researchers have added enzymatic reactions to their artificial intelligence–powered synthesis-planning software RXN. The algorithm can work backward to propose a series of reactions that use enzymes instead of classical chemical catalysts to arrive at a target molecule ( Nat. Commun. 2022, DOI: 10.1038/s41467-022-28536-w ). Using biocatalysts could mean faster, greener reactions. IBM initially launched the software, RXN for Chemistry, to predict the outcomes of chemical reactions. It has since been updated to let users draw a molecule and ask the program to suggest synthetic routes to produce it—a process called retrosynthesis. Accurately forecasting the products of biotransformations has been a stumbling block for retrosynthetic planning, however. The IBM team got around this with its new approach, training the algorithm on four biochemical reaction databases. RXN has about a 40% success rate of planning a synthesis that uses enzymes, and the system has pointed out some errors in the databases