Modeling Epoxidation of Drug-like Molecules with a Deep Machine Learning Network
importance 5/10
partially solved
Computational Drug Discovery Methods drug safety stage: research
Problem
Predicting the site of epoxidation on drug-like molecules, since epoxide reactive metabolites cause drug toxicity.
Why it matters
Identifying epoxidation sites helps interpret adverse drug events and guide design of safer drugs.
What the AI did
A deep machine learning network based on the XenoSite algorithm was trained on 702 epoxidation reactions to predict sites of epoxidation.
Results
Model 'accurately predicted sites of epoxidation'; no quantitative metrics stated
Impact
Aids prediction of reactive metabolite formation for drug safety assessment.
Record
| Venue | ACS Central Science |
|---|---|
| Published | 2015 |
| Work type | article |
| Authors | Tyler B. Hughes, Grover Paul Miller, S. Joshua Swamidass |
| Citations | 203 (OpenAlex, at import time) |
| References | 69 |
| Model or method | XenoSite-based deep machine learning network |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1021/acscentsci.5b00131 |
| Open access | diamond |
| Keywords | epoxidation, reactive metabolites, deep learning, XenoSite, drug safety |
Importance rating
5
real but narrower advance
Abstract links epoxide reactive metabolites to drug toxicity and safer drug design, a real but narrow advance within pharmaceutical safety assessment; no scale stated.