A Full-Spectrum Generative Lead Discovery (FSGLD) pipeline via DRUG-GAN: A multiscale method for drug-like/target-specific compound library generation
Science Advances · 2026
DOI: 10.1126/sciadv.aei5257 ↗Kurz gefasst
Fachartikel in der Zeitschrift Science Advances, 2026. Verfasst von Beihong Ji und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pharmaceutical Sciences and Computational Chemical Genomics Screening Center, School of Pharmacy, University of Pittsburgh, Pittsburgh,…. Hauptthemen laut MEDLINE: Drug Discovery, Small Molecule Libraries. Die Arbeit nennt 120 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, National Science Foundation, National Institutes of Health.
Abstract (Originalsprache)
We present the Full-Spectrum Generative Lead Discovery (FSGLD), a deep learning-driven pipeline for efficient drug lead identification. FSGLD integrates generative modeling with molecular docking, molecular dynamics simulations, ligand-residue interaction profile, MM-PBSA, thermodynamic integration (TI), and experimental validation to bridge theoretical design and practical application. The core multiscale DRUG-GAN models enable generative model-guided discovery of drug-like and target-specific compounds across three scenarios: (i) generation of random drug-like compounds, (ii) generation of target-specific compounds, and (iii) generation of target-specific compound series featuring shared chemical structures. In our assessment across three pharmacologically distinct target classes, FSGLD demonstrates encouraging performance in identifying target-specific candidate compounds. In addition, a computational protocol for TI calculations was established to reduce computation time by 80 to 90% while maintaining accuracy. By integrating generative models with in silico and in vitro evaluation techniques, FSGLD reduces the cost of identifying previously uncharacterized yet viable lead compounds, offering remarkable benefits to both academic and industry.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Pharmaceutical Sciences and Computational Chemical Genomics Screening Center, School of Pharmacy, University of Pittsburgh, Pittsburgh, PA 15261, USA.
- Hangzhou Institute of Medicine, Chinese Academy of Sciences, Hangzhou, 310018 Zhejiang, China.
- School of Medicine, Shanghai University, 200444 Shanghai, China.
ORCID-Profile: Beihong Ji, Yuemin Bian, Xibing He, Haocheng Tang, Junmei Wang
Förderung
- NIGMS NIH HHS
- National Science Foundation
- National Institutes of Health
Zitieren
APA
Ji, B., Brock, M., Wu, Y., Bian, Y., He, X., Liu, Y., Hao, Y., Tang, H., & Wang, J. (2026). A Full-Spectrum Generative Lead Discovery (FSGLD) pipeline via DRUG-GAN: A multiscale method for drug-like/target-specific compound library generation. Science Advances. https://doi.org/10.1126/sciadv.aei5257
Vancouver
Ji B, Brock M, Wu Y, Bian Y, He X, Liu Y, et al. A Full-Spectrum Generative Lead Discovery (FSGLD) pipeline via DRUG-GAN: A multiscale method for drug-like/target-specific compound library generation. Science Advances. 2026. doi:10.1126/sciadv.aei5257
BibTeX
@article{Ji2026,
author = {Ji, Beihong and Brock, Matthew and Wu, Yuhui and Bian, Yuemin and He, Xibing and Liu, Yue and Hao, Yixuan and Tang, Haocheng and Wang, Junmei},
title = {A Full-Spectrum Generative Lead Discovery (FSGLD) pipeline via DRUG-GAN: A multiscale method for drug-like/target-specific compound library generation},
journal = {Science Advances},
year = {2026},
doi = {10.1126/sciadv.aei5257}
}
EUVIMED