Recent advances in molecular representation methods and their applications in scaffold hopping
The rapid evolution of molecular representation methods has significantly advanced the drug discovery process. Advances in language models, graph-based representations, and novel learning strategies have greatly improved the ability to characterize molecules. These AI-driven strategies extend beyond traditional structural data, facilitating exploration of broader chemical spaces and accelerating scaffold hopping. This review summarizes key advancements, discusses their advantages over conventional techniques, and highlights challenges in data quality and real-world applications.
Authors
- Sanan Wu
- Xiangcheng Li (ORCID: https://orcid.org/0000-0002-3909-6316)
- Pengxuan Ren (ORCID: https://orcid.org/0000-0001-6475-3974)
- Fang Bai (ORCID: https://orcid.org/0000-0003-1468-5568)
- Chao Xu (ORCID: https://orcid.org/0000-0001-8970-1935)
- Lu Liu (ORCID: https://orcid.org/0000-0001-5101-5067)
- Lin Wang (ORCID: https://orcid.org/0000-0003-2482-7638)
- Xinyue Ma (ORCID: https://orcid.org/0000-0003-3463-9809)
- Fang Cai (ORCID: https://orcid.org/0000-0003-1002-7146)
- Shihang Wang (ORCID: https://orcid.org/0000-0002-5753-6450)
- Ran Zhang (ORCID: https://orcid.org/0000-0001-6130-8349)
- Q. F. Qian
- Yilin Tang (ORCID: https://orcid.org/0000-0001-8233-2401)
Institutions
- Chinese Academy of Medical Sciences & Peking Union Medical College (CN)
- ShanghaiTech University (CN)
- Suzhou Institute of Systems Medicine (CN)
- Shanghai Clinical Research Center (CN)
Publication Details
- Journal
- npj Drug Discovery.
- Published
- 2025-06-28
- DOI
- https://doi.org/10.1038/s44386-025-00017-2
- Citations
- 13
- Primary Topic
- Computational Drug Discovery Methods
- Type
- article
- Field-Weighted Citation Impact
- 19.56
Funders
- National Natural Science Foundation of China
- ShanghaiTech University
- National Key Research and Development Program of China