EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

DeepDegradome: A structure-aware deep learning framework for PROTAC and ligand generation against protein targets

Qiaoyu Hu iD, Yu Cao, PengXuan Ren, Xianglei Zhang, Fenglei Li, Xueyuan Zhang, Fengyu Cai, Ran Zhang, Yongqi Zhou, Lianghe Mei iD, Fang Bai iD

Proceedings of the National Academy of Sciences · 17.03.2026 · 123 (11)

DOI: 10.1073/pnas.2518248123 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, erschienen am 17.03.2026. Verfasst von Qiaoyu Hu und 10 weiteren Autor:innen. Beteiligte Einrichtung: Shanghai Institute for Advanced Immunochemical Studies, ShanghaiTech University, Shanghai 201210, China. Hauptthemen laut MEDLINE: Deep Learning, Drug Discovery, Proteins. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Shanghai Science and Technology Development Foundation (Shanghai Science and Technology Foundation), ShanghaiTech AI4S Initiative, MOST | National Key Research and Development Program of China (NKPs).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Targeted protein degradation is a promising strategy for drug discovery, but designing effective PROTACs remains challenging, especially for proteins without well-defined binding sites. Current methods rely on modifying linkers between fixed ligands, which limits the diversity and innovation of the overall molecular architecture of PROTAC. Here, we introduce DeepDegradome, an AI-powered method that automates the structure-aware design of both small-molecule ligands and PROTACs. It employs a large fragment library constructed from public databases and applies an in-house docking method (iFitDock) to obtain initial binding fragments. DeepDegradome builds ligands by assembling these fragments based on the shape and physicochemical features of the target protein pocket. It can further construct PROTACs from these generated ligands, eliminating the dependency on predefined warheads or E3 ligands. Compared to other AI models, DeepDegradome produces more valid, drug-like molecules with higher predicted binding affinity. We demonstrate DeepDegradome’s effectiveness by designing and validating multiple potency inhibitors and PROTACs for two protein targets: WDR5 and CDK9. One synthesized compound showed excellent agreement between predicted and actual binding conformation confirmed by X-ray crystallography. By combining ligand and PROTAC design in one system, DeepDegradome offers a scalable and reliable tool for discovering new drugs against protein targets.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningDrug DiscoveryProteinsProteolysis Targeting ChimeraLigandsBinding SitesDrug DesignMolecular Docking SimulationProtein BindingHumansSmall Molecule LibrariesProteolysis

Stichwörter der Autor:innen

PROTACsdeep learningfragment-based drug designmolecular generation

Substanzen

Proteolysis Targeting ChimeraLigandsProteinsSmall Molecule Libraries

Autor:innen und Institutionen

  • Shanghai Institute for Advanced Immunochemical Studies, ShanghaiTech University, Shanghai 201210, China.
  • School of Life Science and Technology, ShanghaiTech University, Shanghai 201210, China.
  • School of Pharmacy, Shanxi Medical University, Taiyuan 030607, China.
  • School of Information Science and Technology, ShanghaiTech University, Shanghai 201210, China.
  • Suzhou Institute of Materia Medica, Chinese Academy of Sciences, Suzhou 215004, China.
  • Shanghai Clinical Research and Trial Center, Shanghai 201210, China.

ORCID-Profile: Qiaoyu Hu, Lianghe Mei, Fang Bai

Förderung

  • Shanghai Science and Technology Development Foundation (Shanghai Science and Technology Foundation)
  • ShanghaiTech AI4S Initiative
  • MOST | National Key Research and Development Program of China (NKPs)
  • Fundamental Research Program of Shanxi Province

Zitieren

APA

Hu, Q., Cao, Y., Ren, P., Zhang, X., Li, F., Zhang, X., Cai, F., Zhang, R., Zhou, Y., Mei, L., & Bai, F. (2026). DeepDegradome: A structure-aware deep learning framework for PROTAC and ligand generation against protein targets. Proceedings of the National Academy of Sciences. https://doi.org/10.1073/pnas.2518248123

Vancouver

Hu Q, Cao Y, Ren P, Zhang X, Li F, Zhang X, et al. DeepDegradome: A structure-aware deep learning framework for PROTAC and ligand generation against protein targets. Proceedings of the National Academy of Sciences. 2026. doi:10.1073/pnas.2518248123

BibTeX
@article{Hu2026,
  author = {Hu, Qiaoyu and Cao, Yu and Ren, PengXuan and Zhang, Xianglei and Li, Fenglei and Zhang, Xueyuan and Cai, Fengyu and Zhang, Ran and Zhou, Yongqi and Mei, Lianghe and Bai, Fang},
  title = {DeepDegradome: A structure-aware deep learning framework for PROTAC and ligand generation against protein targets},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2026},
  doi = {10.1073/pnas.2518248123}
}
Kostenlos starten →