Work

Screening of Natural Product-Derived USP7 Inhibitors for Cancer Therapy via Integrated Machine Learning and Molecular Simulations

Year: 2026

Type: article

Abstract: Ubiquitination, a crucial cellular protein regulation process, is linked to various diseases, including cancer. Deubiquitinases (DUBs) can reverse ubiquitination, offering a therapeutic strategy. USP7, a DUB, is a key target in oncology due to its role in destabilizing p53, and small-molecule inhibitors could restore p53 activity and combat tumor growth. In this study, we integrated a machine learning (ML)-based screening approach with molecular docking and molecular dynamics (MD) simulations in order to identify potential small-molecule inhibitors of USP7. ML-based screening identified 22 active molecules from a library of 2301 natural compounds. Among the 22 active compounds, only fifteen compounds fulfilled the drug-likeness criteria. Subsequently, molecular docking found three compounds, PubChem 162957515, 114917, and 442879 as potential inhibitors based on binding affinity and interactions. Further, MD simulations and MM-PBSA analyses were performed to evaluate the stability and dynamic behavior of the complexes. Binding energy calculations Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA) revealed that compounds PubChem 114917 and 162957515 exhibited strong binding affinities of -20.98 kcal/mol and -18.68 kcal/mol, respectively, implying that these compounds could serve as promising inhibitors for the development of anticancer therapeutics.

Source: Current Issues in Molecular Biology

DOI: 10.3390/cimb48060621

Author: Faris Alrumaihi

Institution: Qassim University

Language: English

Topic: Computational Drug Discovery Methods

Subfield: Computational Theory and Mathematics

Field: Computer Science

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $2,654