Deep learning mines antimicrobial peptides from the human gut microbiome
Source check: Needs source check. Not independent scientific confirmation.
Finding: An NLP-style pipeline (LSTM, attention, BERT) flagged 2,349 candidate antimicrobial peptides in gut microbiome data; 216 were synthesized and 181 showed activity, and the 11 most potent were effective against resistant Gram-negative pathogens, including a mouse lung infection model. Abstract-level review only; clinical relevance untested.
Technology: LSTM, attention and BERT neural pipeline
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- Neural language models ranked candidate AMPs from metagenome sequences; wet-lab tests validated them.
- Test and result
- 216 candidates synthesized, 181 active (>83%); 11 potent peptides tested; mouse bacterial lung infection model.
- Limitation / still unknown
- Toxicity, pharmacokinetics and human efficacy not assessed in the abstract; full text not read.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://europepmc.org/search?query=DOI%3A10.1038/s41587-022-01226-0
Needs source check · editorial details
The source has not been fully checked or contains conflicting information. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review scope and date are not recorded; source checked does not establish full-text review.
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.