2. Academic Validation
  3. Accelerating discovery of bioactive ligands with pharmacophore-informed generative models

Accelerating discovery of bioactive ligands with pharmacophore-informed generative models

  • Nat Commun. 2025 Mar 10;16(1):2391. doi: 10.1038/s41467-025-56349-0.
Weixin Xie # 1,  Jianhang Zhang # 2,  Qin Xie 2,  Chaojun Gong 2,  Yuhao Ren 3,  Jin Xie 1,  Qi Sun 3 4 5,  Youjun Xu 6,  Luhua Lai 7 8 9 10,  Jianfeng Pei 11 12
Affiliations

Affiliations

  • 1 Center for Quantitative Biology, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China.
  • 2 Infinite Intelligence Pharma, Beijing, China.
  • 3 BNLMS, Peking-Tsinghua Center for Life Sciences at the College of Chemistry and Molecular Engineering, Peking University, Beijing, China.
  • 4 Peking University Chengdu Academy for Advanced Interdisciplinary Biotechnologies, Chengdu, China.
  • 5 Research Unit of Drug Design Method, Chinese Academy of Medical Sciences, Beijing, China.
  • 6 Infinite Intelligence Pharma, Beijing, China. xuyj@iipharma.cn.
  • 7 Center for Quantitative Biology, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China. lhlai@pku.edu.cn.
  • 8 BNLMS, Peking-Tsinghua Center for Life Sciences at the College of Chemistry and Molecular Engineering, Peking University, Beijing, China. lhlai@pku.edu.cn.
  • 9 Peking University Chengdu Academy for Advanced Interdisciplinary Biotechnologies, Chengdu, China. lhlai@pku.edu.cn.
  • 10 Research Unit of Drug Design Method, Chinese Academy of Medical Sciences, Beijing, China. lhlai@pku.edu.cn.
  • 11 Center for Quantitative Biology, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China. jfpei@pku.edu.cn.
  • 12 Research Unit of Drug Design Method, Chinese Academy of Medical Sciences, Beijing, China. jfpei@pku.edu.cn.
  • # Contributed equally.
PMID: 40064886 DOI: 10.1038/s41467-025-56349-0
Abstract

Deep generative models have advanced drug discovery but often generate compounds with limited structural novelty, providing constrained inspiration for medicinal chemists. To address this, we develop TransPharmer, a generative model that integrates ligand-based interpretable pharmacophore fingerprints with a generative pre-training transformer (GPT)-based framework for de novo molecule generation. TransPharmer excels in unconditioned distribution learning, de novo generation, and scaffold elaboration under pharmacophoric constraints. Its unique exploration mode could enhance scaffold hopping, producing structurally distinct but pharmaceutically related compounds. Its efficacy is validated through two case studies involving the Dopamine Receptor D2 (DRD2) and polo-like kinase 1 (PLK1). Notably, three out of four synthesized PLK1-targeting compounds show submicromolar activities, with the most potent, IIP0943, exhibiting a potency of 5.1 nM. Featuring a new 4-(benzo[b]thiophen-7-yloxy)pyrimidine scaffold, IIP0943 also has high PLK1 selectivity and submicromolar inhibitory activity in HCT116 cell proliferation. TransPharmer offers a promising tool for discovering structurally novel and bioactive ligands.

Figures
Products