Related Experiment Video
Updated: Aug 14, 2026
10:29
Quantitative Structure-Activity Relationship, Activity Prediction, and Molecular Dynamics of Non-nucleotide Reverse Transcriptase Inhibitors
Published on: May 9, 2025
Discovery of a potent TDP1 inhibitor through machine learning-driven predictive modeling combined with
Huang Zeng1, Manyi Zhang1, Bo Qiu1
1Medical College, Jiaying University Meizhou 514031 China 202101261@jyu.edu.cn niehua@jyu.edu.cn.
RSC Advances
|August 13, 2026
Summary
Researchers developed a computational framework to discover Tyrosyl-DNA phosphodiesterase I (TDP1) inhibitors, identifying a potent new compound, AO65. This approach integrates machine learning and structure-based docking for anticancer drug discovery.
Area of Science:
- Medicinal Chemistry
- Computational Drug Discovery
- Biochemistry
Background:
- Tyrosyl-DNA phosphodiesterase I (TDP1) is a critical enzyme in DNA repair and a promising anticancer target.
- Developing potent and drug-like TDP1 inhibitors is challenging due to limited structural diversity of known compounds.
Purpose of the Study:
- To develop an integrated computational framework for discovering novel TDP1 inhibitors.
- To identify and validate potent TDP1 inhibitors using a combination of machine learning and structure-based methods.
Main Methods:
- Assembled a dataset of 2040 compounds for training binary classification models using six machine learning algorithms and a deep neural network.
- Employed five molecular fingerprint representations and five graph neural network architectures.
- Validated the best performing SVM::RDKitDes model (AUC=0.89) and applied it to screen over 200,000 compounds.
- Conducted drug-likeness filtering, hierarchical docking, and biological evaluation of selected candidates.
Main Results:
- The SVM::RDKitDes model demonstrated high predictive performance (AUC=0.89, F1=0.78, BA=0.80).
- SHAP analysis identified 20 key descriptors for TDP1 inhibition.
- Screening identified 16 potential candidates, with compound AO65 showing potent TDP1 inhibition (IC50 = 0.80 ± 0.02 µM).
- Quantum chemical calculations and docking elucidated AO65's binding mode within the TDP1 catalytic domain.
Conclusions:
- The integrated ML and structure-based approach is effective for discovering novel TDP1 inhibitors.
- Compound AO65 represents a promising lead compound for further development in TDP1-targeted anticancer therapies.
