Machine learning-powered antibiotics phenotypic drug discovery
Problem
Identifying novel antibiotics and determining compound mode of action remains a major drug-discovery challenge, with weak hits often underexploited.
Why it matters
New antibiotics are urgently needed as resistance grows and discovery pipelines stall.
What the AI did
Machine learning analysis of bacterial phenotypic fingerprint (BPF) profiles narrowed down, compared and predicted compound mode of action, including weak antibacterial hits.
Results
Predicted compound mode of action and guided SAR optimization across species; no quantitative metrics stated in source
Impact
Evidence supports ML-guided mode-of-action prediction, but no quantitative benchmarks are provided.
Record
| Venue | Scientific Reports |
|---|---|
| Published | 2019 |
| Work type | article |
| Authors | Sannah Zoffmann, Maarten Vercruysse, Fethallah Benmansour, Andreas Maunz, Luise Wolf, Rita Blum Marti, Tobias Heckel, Haiyuan Ding et al. |
| Citations | 111 (OpenAlex, at import time) |
| References | 29 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41598-019-39387-9 |
| Open access | gold |
| Keywords | antibiotics, machine learning, mode of action, phenotypic screening, drug discovery |
Importance rating
significant — thousands to millions affected
Abstract states identifying novel antibiotics remains a major challenge for drug discovery; approach widens antibacterial chemical space of pharmaceutical libraries and extends to broader phenotypic drug discovery.