AI Science Tracker

ApexAmphion designs antibiotic peptides: 100/100 designed peptides active in vitro

University of Pennsylvania; Stanford University (affiliations to verify) · 23 Sep 2025 · medicine

Discovery

ApexAmphion designs antibiotic peptides: 100/100 designed peptides active in vitro

Source check: Needs source check. Not independent scientific confirmation.

Finding: 100 designed peptides all showed low MIC (nanomolar for some); 99 of 100 were broad-spectrum against at least two clinically relevant bacteria; lead molecules acted on the cytoplasmic membrane. Preprint; in vitro only; no in vivo or toxicity assessment in the abstract; affiliations unverified.

Technology: 6.4B-parameter protein language model fine-tuned with PPO reinforcement learning and MIC classifier reward

Publication: Preprint · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
Language model plus RL generated peptides optimised against a learned MIC classifier.
Test and result
100 designed peptides all showed low MIC (nanomolar for some); 99 of 100 were broad-spectrum against at least two clinically relevant bacteria; lead molecules acted on the cytoplasmic membrane.
Limitation / still unknown
Preprint; in vitro only; no in vivo or toxicity assessment in the abstract; affiliations unverified.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=DOI:%2210.1101/2025.09.23.678086%22&resultType=core&format=json

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: linked separately.

Review date: 17 Nov 2026. Review scope: Abstract and metadata via Europe PMC; full text not read.

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 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse medicine discoveries.

Related catalog entries

Links use documented relationships where available, plus shared catalog context. A shared topic or model family does not establish a scientific dependency.