The paper titled “multi-omics pan-cancer profiling of CDK2 and in silico identification of plant-derived inhibitors using machine learning approaches” conducts a comprehensive multi-omics analysis of CDK2 across many cancer types. The authors evaluate CDK2 expression levels, prognostic significance, genetic alterations, and immune-infiltration patterns across tumors — showing that in multiple cancers CDK2 is significantly overexpressed and in some cases correlated with poor overall and disease-free survival, indicating CDK2’s viability as a context-dependent prognostic biomarker. In addition, the study uses cheminformatics and machine-learning to screen phytocompounds from several medicinal plants, then carries out molecular docking, pharmacokinetics (ADME/T), and MD-based dynamics & free energy (MM-GBSA) calculations to identify potential plant-derived inhibitors of CDK2. Among the top hits, compounds such as Withanolide M, Withanolide K, and Ergosterol showed strong binding affinities (docking scores −10.2 to −9.9 kcal/mol), favorable predicted pharmacokinetic properties, and low toxicity risk — comparable or better than the reference CDK2 inhibitor used (RSC Adv., 2025,15, 36938-36968). Their MD simulations and MM-GBSA analysis supported stable binding over time, consistent with docking predictions. In short: the paper (i) establishes CDK2 as an important pan-cancer target from multi-omics data, (ii) identifies promising natural phytochemicals as CDK2 inhibitors, and (iii) provides in silico evidence (docking + ADMET + dynamics) that these compounds could serve as leads for further experimental validation. Since your work already involves natural phytochemicals and cell-cycle regulators (CDK1, AURKA, CDC20), this paper could be a very relevant reference — maybe I can help you write a proper citation paragraph or place its findings in context with your review article, if you like.
To read more about the article click Here
Source: RSC Adv., 2025,15, 36938-36968
