Interpretable prediction of aggregation-induced emission molecules based on graph neural networks
We developed an interpretable graph neural network (96.4% accuracy) for AIEgen identification, revealing 24 characteristic functional groups. Based on these insights, two virtual library strategies (self-fragment and donor-acceptor docking) were proposed and predicted four experimentally confirmed AIEgens successfully, which establishes a rational design framework for AIE materials.
Authors
- Xiaoyan Zheng (ORCID: https://orcid.org/0000-0002-2997-9756)
- Huanxiao Mai (ORCID: https://orcid.org/0000-0003-0682-9189)
- Dong Wang (ORCID: https://orcid.org/0000-0001-5137-0771)
- Yi Zeng (ORCID: https://orcid.org/0000-0002-6901-9194)
- Shichen Zhang (ORCID: https://orcid.org/0009-0007-1973-306X)
- Jun Zhu (ORCID: https://orcid.org/0000-0003-1186-4392)
Institutions
- Shenzhen University (CN)
- Ministry of Industry and Information Technology (CN)
Publication Details
- Journal
- Chemical Communications
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1039/d5cc01949d
- Citations
- 2
- Primary Topic
- Luminescence and Fluorescent Materials
- Type
- article
- Field-Weighted Citation Impact
- 1.04
Funders
- National Natural Science Foundation of China
- Beijing Municipal Natural Science Foundation