---
id: "2023-12-20-ai-new-antibiotic-structural-class"
url: "https://postcutoff.com/e/2023-12-20-ai-new-antibiotic-structural-class/"
as_of: "2026-11-13T13:49:00+01:00"
date: "2023-12-20"
date_precision: day
category: science
importance: 3
confidence: high
status: [Result confirmed]
verification: Peer-reviewed
sources: 2
editor: human editor
human_review: null
version: "2026-10-04"
---

As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/e/2023-12-20-ai-new-antibiotic-structural-class/

# Explainable deep learning discovers a new structural class of antibiotics against MRSA

Felix Wong, James Collins and colleagues (Nature, Dec 2023) screened ~39,000 compounds, trained graph neural networks, and used explainable substructure analysis on ~12M compounds. They found a new structural class of antibiotics active against MRSA and VRE that worked in mouse models.

## Key facts
- Nature, published online 20 Dec 2023
- ~39,000 compounds tested experimentally; ~12M scored computationally
- Active against MRSA and vancomycin-resistant enterococci; effective topically and systemically in mice

## What happened

Rather than a black-box ranking, the model identified which chemical substructures drove predicted activity, leading chemists to a new antibiotic class.

## Why it matters

New structural classes of antibiotics are rarely discovered. This one came from interpretable AI, which also showed chemists why the molecules work.

## Your AI and this story

- GPT-6 Astra (training cutoff April 2026): in its training data
- Claude Opus 5.5 (training cutoff June 2026): in its training data
- Gemini 4 Hydrogen (training cutoff March 2026): in its training data
- Grok 4.7 (training cutoff May 2026): in its training data

## Sources

1. [Discovery of a structural class of antibiotics with explainable deep learning (Nature)](https://www.nature.com/articles/s41586-023-06887-8) (nature.com, paper)
2. [Broad Institute: Researchers use AI to identify new class of antibiotic candidates](https://www.broadinstitute.org/news/researchers-use-ai-identify-new-class-antibiotic-candidates) (broadinstitute.org, press)

## Changes

- 2026-09-29 (filed): Created

## Related

- 2020-02-20: [Deep learning discovers halicin, a structurally new broad-spectrum antibiotic](https://postcutoff.com/e/2020-02-20-halicin-ai-antibiotic/index.md)