Journal article
A White-Box Machine Learning Approach for Revealing Antibiotic Mechanisms of Action
Massachusetts Institute of Technology1
Harvard University2
Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark3
Network Reconstruction in Silico Biology, Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark4
iLoop, Translational Management, Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark5
Boston University6
Big Data 2 Knowledge, Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark7
Current machine learning techniques enable robust association of biological signals with measured phenotypes, but these approaches are incapable of identifying causal relationships. Here, we develop an integrated "white-box" biochemical screening, network modeling, and machine learning approach for revealing causal mechanisms and apply this approach to understanding antibiotic efficacy.
We counter-screen diverse metabolites against bactericidal antibiotics in Escherichia coli and simulate their corresponding metabolic states using a genome-scale metabolic network model. Regression of the measured screening data on model simulations reveals that purine biosynthesis participates in antibiotic lethality, which we validate experimentally.
We show that antibiotic-induced adenine limitation increases ATP demand, which elevates central carbon metabolism activity and oxygen consumption, enhancing the killing effects of antibiotics. This work demonstrates how prospective network modeling can couple with machine learning to identify complex causal mechanisms underlying drug efficacy.
Language: | English |
---|---|
Year: | 2019 |
Pages: | 1649-1661.e9 |
ISSN: | 10974172 and 00928674 |
Types: | Journal article |
DOI: | 10.1016/j.cell.2019.04.016 |
ORCIDs: | Schrübbers, Lars and Palsson, Bernhard O |
ATP Adenylate energy charge Antibiotics Biochemical screen LC-MS/MS Machine learning Metabolism NADPH:NADP+ ratio Network modeling Purine biosynthesis
Adenine Anti-Bacterial Agents Computational Biology Drug Evaluation, Preclinical Escherichia coli Machine Learning Metabolic Networks and Pathways Models, Theoretical NADPH:NADP(+) ratio Purines adenylate energy charge antibiotics biochemical screen machine learning metabolism network modeling purine biosynthesis