AMCL: supervised contrastive learning with hard sample mining for multi-functional therapeutic peptide prediction
Comprehensive experiments show that AMCL significantly outperforms existing methods across multiple key metrics, including Absolute true, Accuracy, Macro-F1, and Micro-F1, establishing a new state-of-the-art in therapeutic peptide multi-functional prediction.
Authors
- Jintao Zhao (ORCID: https://orcid.org/0000-0002-7328-5401)
- Junfeng Xia (ORCID: https://orcid.org/0000-0003-3024-1705)
- Jiwei Fang (ORCID: https://orcid.org/0000-0003-2740-3187)
- Jianping Zhao (ORCID: https://orcid.org/0000-0002-4080-8448)
- Henghui Fan (ORCID: https://orcid.org/0000-0002-8150-9954)
Institutions
- Anhui University (CN)
- Hefei Institutes of Physical Science (CN)
- Xinjiang University (CN)
Publication Details
- Journal
- BMC Biology
- Published
- 2025-07-01
- DOI
- https://doi.org/10.1186/s12915-025-02273-0
- Citations
- 1
- Primary Topic
- Chemical Synthesis and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.81
Funders
- National Natural Science Foundation of China
- National Key Research and Development Program of China