Deep learning guides discovery of the STB-C017 kynurenine-pathway inhibitor
Source check: Source checked. Not independent scientific confirmation.
Finding: Convolutional neural-network models trained on prior IDO and TDO bioactivity guided screening and assay-driven refinement, leading to STB-C017. The study reports dual inhibition in biochemical and cell assays, reduced kynurenine in mice and enhanced antitumor responses when combined with immune-checkpoint inhibitors in mouse models. This entry represents the November 2021 journal version; an earlier 2021 Research Square manuscript concerns the same study, and efficacy or safety in patients is not established.
Technology: Convolutional neural-network bioactivity classifiers
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- Separate CNN classifiers ranked potential IDO and TDO inhibitors. Four cycles of predictions, selection and measured assays updated the models before lead selection.
- Test and result
- Biochemical and cell assays tested IDO/TDO inhibition, followed by mouse studies of kynurenine suppression and antitumor combinations in the originating study.
- Limitation / still unknown
- Clinical efficacy and safety are not established; originating-study assays and mouse experiments are not independent replication.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.1080/2162402X.2021.2005280 · https://pmc.ncbi.nlm.nih.gov/articles/PMC8632076/
Source checked · editorial details
The source page was checked against this entry. 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 date: 21 Nov 2026. Review scope: Primary full article Methods, result-specific experiments, discussion, affiliations and journal publication metadata; no independent replication performed..
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.