Multitask neural screening identifies compounds active against Leishmania infantum in vitro
Source check: Source checked. Not independent scientific confirmation.
Finding: Multitask neural classification and regression screened approximately 1.3 million purchasable compounds for Leishmania infantum activity. After physicochemical and human structural filtering, 20 selected compounds were tested; nine inhibited promastigotes in vitro. Three had more favorable source-reported selectivity, with LC-15 showing an IC50 of 1.05 µM and selectivity index 30.9. LC-15 was already disclosed in a 2023 antitrypanosomal study; this entry represents the Leishmania result, not a new chemical entity. Intracellular parasite activity, animal efficacy and a proposed proteasome mechanism were not established.
Technology: Multitask deep neural network classification and regression; molecular fingerprints
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- Multitask DNN classifiers and regressors transferred information across three Leishmania species to prioritize untested molecules; physicochemical filters, similarity analysis and human inspection then selected 20 for experiments.
- Test and result
- Twenty model-prioritized, subsequently filtered compounds were evaluated against Leishmania infantum promastigotes. Nine were active; cytotoxicity and selectivity measurements were reported for selected active hits. Three had source-described selectivity indices above 9.
- Limitation / still unknown
- No intracellular-amastigote efficacy, animal therapeutic efficacy or clinical benefit is established. Six active compounds had less favorable selectivity. LC-15 is the same chemical entity disclosed in the 2023 antitrypanosomal study, whose supplement also reports the same MRC-5 cytotoxicity value; independently new cytotoxicity evidence is not inferred. Its proteasome mechanism remains a hypothesis, and the observed hit rate is not a controlled comparison with conventional screening.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.1021/acsomega.4c07994 · https://pmc.ncbi.nlm.nih.gov/articles/PMC11696749/ · https://doi.org/10.4155/fmc-2023-0074 · https://doi.org/10.25402/FMC.24131556.v1 · https://doi.org/10.1021/acsomega.4c07994.s001
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: Corrected exact primary DOI/title/PMCID and electronic date; full computational prioritization and result-specific activity/selectivity paragraphs; current PubMed publication-status relationships. Earlier metadata lookup had retrieved a different paper and is not used as evidence. Compared the original 2023 and 2024 author supplements by exact LC-15 chemical name, structure, supplier identifier and NMR evidence; the represented novelty is the Leishmania assay, not chemical-entity novelty..
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 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.