An interpretable framework applying protein words to predict protein–small molecule complementary pairing rules
- 1MOE Key Laboratory of Bioinformatics, State Key Laboratory of Molecular Oncology, Beijing Frontier Research Center for Biological Structure, School of Pharmaceutical Sciences, Tsinghua University Beijing 100084 China
Abstract
Despite the high accuracy of ‘black box’ deep learning models, drug discovery still relies on protein–ligand interaction principles and heuristics. To improve interpretability of protein–small molecule binding predictions, we developed the PWRules framework, which applies binding affinity data to identify privileged small molecule fragments and subsequently defines complementary pairing rules between these fragments and protein words (semantic sequence units) through an interpretability module. The resulting word–fragment rules are then used by the PWScore function to prioritize active compounds. Evaluations on benchmark datasets show that PWScore achieves competitive performance comparable to the physics-based model (Glide) and the deep learning model (PSICHIC) and shows broad applicability for protein targets outside the training dataset, e.g., SARS-CoV-2 main protease. Notably, PWScore captures complementary interaction information, yielding superior enrichment performance when integrated with these established methods. Structural analysis of protein–ligand complexes indicates that learned word–fragment rules are significantly enriched near ligand-binding pockets, despite training without explicit structural guidance. By extracting and applying complementary pairing rules, PWRules provides an interpretable framework for drug discovery.
Related
- DISSOLVR: An Interpretable and Fast Framework for Aqueous and Organic Solubility Prediction
- StructRSMA: A Proof-of-Concept Framework for PDB-Derived Contact-Supervised Transfer in RNA-Small Molecule Binding Affinity Prediction
- Chemistry-infused bandit framework for interpretable autonomous campaign of electrocatalysis
- DeepHeal: An Interpretable Framework for Phenotype-Based Antineoplastic Drug Discovery Using Drug-Perturbation Proteomics
- Interpretable Machine Learning Framework for Nb─Si Based Alloy Design with Enhanced Fracture Toughness
BibTeX
@article{chen2026interpretable,
title = {{An interpretable framework applying protein words to predict protein–small molecule complementary pairing rules}},
author = {Jingke Chen and Jingrui Zhong and Tazneen Hossain Tani and Zidong Su and Xiaochun Zhang and Boxue Tian},
journal = {Chemical Science},
year = {2026},
month = sep,
doi = {10.1039/d6sc03365b},
url = {https://doi.org/10.1039/d6sc03365b},
note = {CC BY-NC 3.0},
}
RIS
TY - JOUR
TI - An interpretable framework applying protein words to predict protein–small molecule complementary pairing rules
AU - Jingke Chen
AU - Jingrui Zhong
AU - Tazneen Hossain Tani
AU - Zidong Su
AU - Xiaochun Zhang
AU - Boxue Tian
PY - 2026
DA - 2026/09/08
JO - Chemical Science
DO - 10.1039/d6sc03365b
UR - https://doi.org/10.1039/d6sc03365b
AB - Despite the high accuracy of ‘black box’ deep learning models, drug discovery still relies on protein–ligand interaction principles and heuristics. To improve interpretability of protein–small molecule binding predictions, we developed the PWRules framework, which applies binding affinity data to identify privileged small molecule fragments and subsequently defines complementary pairing rules between these fragments and protein words (semantic sequence units) through an interpretability module. The resulting word–fragment rules are then used by the PWScore function to prioritize active compounds. Evaluations on benchmark datasets show that PWScore achieves competitive performance comparable to the physics-based model (Glide) and the deep learning model (PSICHIC) and shows broad applicability for protein targets outside the training dataset, e.g., SARS-CoV-2 main protease. Notably, PWScore captures complementary interaction information, yielding superior enrichment performance when integrated with these established methods. Structural analysis of protein–ligand complexes indicates that learned word–fragment rules are significantly enriched near ligand-binding pockets, despite training without explicit structural guidance. By extracting and applying complementary pairing rules, PWRules provides an interpretable framework for drug discovery.
SN - 2041-6520
SN - 2041-6539
ER -