An Explicit Interaction-Prompted Diffusion Framework for High-Fidelity 3D Molecular Generation
Journal of the American Chemical Society · 2026
DOI: 10.1021/jacs.6c08189 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the American Chemical Society, 2026. Verfasst von Huabin Du und 9 weiteren Autor:innen. Beteiligte Einrichtung: Chenzhu (MoMeD) Biotechnology Co., Ltd., Hangzhou311100, People's Republic of China. Hauptthemen laut MEDLINE: Drug Design, Proteins. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 64 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China, Chenzhu (MoMeD) Biotechnology Co., Ltd., Drug?Target Interaction Database Development.
Abstract (Originalsprache)
Current structure-based drug design generative models often struggle to faithfully recapitulate genuine ligand–protein binding interactions. Instead, under the coupling of implicit learning architectures and biased training data, they tend to learn spurious statistical correlations. To address this, we propose EIP-Diff (Explicit Interaction-Prompted Diffusion), an architecture featuring a novel explicit interaction-prompt embedding mechanism that is better suited for real-world target-specific drug design. This architecture replaces biased implicit learning with explicit, residue-level biological guidance, thereby promoting more fine-grained geometric fidelity and more precise interaction-aware conditioning. To fully realize the capabilities of EIP-Diff and provide a reliable basis for performance evaluation, we further constructed CrystalData set, which provides higher-fidelity and less-biased structural supervision than existing data sets. This explicit architecture markedly improves distribution consistency: even when trained on the crossdocked data set, EIP-Diff achieves the highest alignment with authentic pharmacological distributions among evaluated models. Training on CrystalData set further enhances this alignment and improves 3D geometric accuracy, while retaining strong controllability, high chemical space coverage, and near-perfect uniqueness. In addition, target-based validation on KAT6A and YTHDC1 confirmed that EIP-Diff accurately recapitulates native-like binding modes. Furthermore, in a real-world drug design task against IDO1, we successfully designed a novel lead compound with nanomolar potency (IC50 = 0.31 nM). These results demonstrate that the EIP-Diff architecture can explicitly leverage experimentally derived structural data and biologically meaningful interaction information for target-specific molecular generation, thereby enabling its effective application to real-world structure-based drug design.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Chenzhu (MoMeD) Biotechnology Co., Ltd., Hangzhou311100, People's Republic of China.
- College of Pharmaceutical Sciences, Zhejiang University, Hangzhou310058, People's Republic of China.
- Zhejiang Provincial People's Hospital, Hangzhou310014, People's Republic of China.
- Key Laboratory of Systems Health Science of Zhejiang Province, School of Life Science, Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou310024, People's Republic of China.
- Department of Chemistry, University of Southern California Los Angeles, Los Angeles, California90089-1062, United States.
ORCID-Profile: Arieh Warshel, Tingjun Hou, Chen Bai
Förderung
- National Natural Science Foundation of China
- National Key Research and Development Program of China
- Chenzhu (MoMeD) Biotechnology Co., Ltd., Drug?Target Interaction Database Development
Zitieren
APA
Du, H., Wang, M., Luo, M., Yang, M., Yan, Y., An, K., Zhu, X., Warshel, A., Hou, T., & Bai, C. (2026). An Explicit Interaction-Prompted Diffusion Framework for High-Fidelity 3D Molecular Generation. Journal of the American Chemical Society. https://doi.org/10.1021/jacs.6c08189
Vancouver
Du H, Wang M, Luo M, Yang M, Yan Y, An K, et al. An Explicit Interaction-Prompted Diffusion Framework for High-Fidelity 3D Molecular Generation. Journal of the American Chemical Society. 2026. doi:10.1021/jacs.6c08189
BibTeX
@article{Du2026,
author = {Du, Huabin and Wang, Mingyang and Luo, Mengqi and Yang, Mengke and Yan, Yingchao and An, Ke and Zhu, Xiaohong and Warshel, Arieh and Hou, Tingjun and Bai, Chen},
title = {An Explicit Interaction-Prompted Diffusion Framework for High-Fidelity 3D Molecular Generation},
journal = {Journal of the American Chemical Society},
year = {2026},
doi = {10.1021/jacs.6c08189}
}
EUVIMED