1. Prioritising CDK2 selectivity rather than potency alone1
In an accepted-paper abstract, researchers report that a machine-learning selectivity margin recovered CDK2-biased compounds better than predicted potency alone, including evaluation on held-out scaffolds. Their workflow also nominated sulfonamide analogues for testing. The comparison is useful for choosing which molecules merit experiments; the candidates remain computational predictions. This mention is based on the publisher abstract, without full-methods assessment.1
References
Precious A. Akinnusi; Gladys D. Egunjobi; Ayomide J. Akinnusi; Fayowole D. Ogunbiyi. A selectivity-aware machine-learning workflow for prioritizing CDK2-biased kinase inhibitor candidates. Scientific Reports; 2026; Peer-reviewed accepted article, early publisher release; final Version of Record pending. DOI: 10.1038/s41598-026-71767-w. Accessed 2026-09-16T17:06:14.000Z.
Source evidence and access
Abstract paragraphs on predicted selectivity margin, retrospective/scaffold-held-out enrichment, and sulfonamide analog filtering; About this article publication date
Evidence paraphrase: Predicted CDK2 selectivity margin recovered measured CDK2-biased compounds more effectively than potency alone; abstract reports retrospective and scaffold-held-out enrichment. Sulfonamide generation/filtering prioritized221 candidates, with docking and rescoring used for further ranking. The abstract describes candidates for experimental evaluation, not prospective experimental validation.
Publisher abstract, author list and bibliographic metadata read. Full methods and supplement not assessed; mention is explicitly abstract-only.
Publication record
Published 2026-09-16.
Sources, selection and claims were checked in an independent AI editorial review, followed by the AI editor's approval. This is not academic peer review.
- 2026-09-16 · Version 02fb5e3e · Viewing this version