Title: Predicting novel substrates for enzymes with minimal experimental effort with active learning

Abstract

Enzymatic substrate promiscuity is more ubiquitous than previously thought, with significant consequences for understanding metabolism and its application to biocatalysis. This realization has given rise to the need for efficient characterization of enzyme promiscuity. Enzyme promiscuity is currently characterized with a limited number of human-selected compounds that may not be representative of the enzyme's versatility. While testing large numbers of compounds may be impractical, computational approaches can exploit existing data to determine the most informative substrates to test next, thereby more thoroughly exploring an enzyme's versatility. Here, to demonstrate this, we used existing studies and tested compounds for four different enzymes, developed support vector machine (SVM) models using these datasets, and selected additional compounds for experiments using an active learning approach. SVMs trained on a chemically diverse set of compounds were discovered to achieve maximum accuracies of similar to 80% using similar to 33% fewer compounds than datasets based on all compounds tested in existing studies. Active learning-selected compounds for testing resolved apparent conflicts in the existing training data, while adding diversity to the dataset. Finally, the application of these algorithms to wide arrays of metabolic enzymes would result in a library of SVMs that can predict high-probability promiscuous enzymaticmore » reactions and could prove a valuable resource for the design of novel metabolic pathways.« less

Authors:

Pertusi, Dante A. ^[1]; Moura, Matthew E. ^[1]; Jeffryes, James G. ^[2]; Prabhu, Siddhant ^[1]; Walters Biggs, Bradley ^[1]; Tyo, Keith E. J. ^[1]

---

+ Show Author Affiliations

1. Northwestern Univ., Evanston, IL (United States). Dept. of Chemical and Biological Engineering
2. Northwestern Univ., Evanston, IL (United States). Dept. of Chemical and Biological Engineering; Argonne National Lab. (ANL), Argonne, IL (United States). Mathematics and Computer Science Division

Publication Date:

Tue Oct 10 00:00:00 EDT 2017

Research Org.:

Argonne National Laboratory (ANL), Argonne, IL (United States)

Sponsoring Org.:

National Science Foundation (NSF); National Institutes of Health (NIH); Bill and Melinda Gates Foundation; USDOE

OSTI Identifier:

1427497

Grant/Contract Number:  

AC02-06CH11357; T32-GM008449-23

Resource Type:

Accepted Manuscript

Journal Name:

Metabolic Engineering

Additional Journal Information:

Journal Volume: 44; Journal Issue: C; Journal ID: ISSN 1096-7176

Publisher:

Elsevier

Country of Publication:

United States

Language:

English

Subject:

97 MATHEMATICS AND COMPUTING; 37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; Active learning; Enzyme promiscuity; Machine learning