Enhancing Multimodal Protein Function Prediction Through Dual-Branch Dynamic Selection with Reconstructive Pre-Training

Luo, Xiaoling; Chen, Peng; Liu, Chengliang; Jin, Xiaopeng; Wen, Jie; Liu, Yumeng; Wang, Junsong

doi:10.24963/ijcai.2025/845

Enhancing Multimodal Protein Function Prediction Through Dual-Branch Dynamic Selection with Reconstructive Pre-Training

Xiaoling Luo, Peng Chen, Chengliang Liu, Xiaopeng Jin, Jie Wen, Yumeng Liu, Junsong Wang

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 7598-7606.

https://doi.org/10.24963/ijcai.2025/845

Keywords

Multidisciplinary Topics and Applications: MTA: Bioinformatics
Machine Learning: ML: Multi-label learning
Machine Learning: ML: Multi-modal learning
Machine Learning: ML: Multi-view learning

Abstract

Multimodal protein features play a crucial role in protein function prediction. However, these features encompass a wide range of information, ranging from structural data and sequence features to protein attributes and interaction networks, making it challenging to decipher their complex interconnections. In this work, we propose a multimodal protein function prediction method (DSRPGO) by utilizing dynamic selection and reconstructive pre-training mechanisms. To acquire complex protein information, we introduce reconstructive pre-training to mine more fine-grained information with low semantic levels. Moreover, we put forward the Bidirectional Interaction Module (BInM) to facilitate interactive learning among multimodal features. Additionally, to address the difficulty of hierarchical multi-label classification in this task, a Dynamic Selection Module (DSM) is designed to select the feature representation that is most conducive to current protein function prediction. Our proposed DSRPGO model improves significantly in BPO, MFO, and CCO on human datasets, thereby outperforming other benchmark models.

Xiaoling Luo, Peng Chen, Chengliang Liu, Xiaopeng Jin, Jie Wen, Yumeng Liu, and Junsong Wang. 2025. Enhancing Multimodal Protein Function Prediction Through Dual-Branch Dynamic Selection with Reconstructive Pre-Training. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 7598–7606. https://doi.org/10.24963/ijcai.2025/845

@inproceedings{ijcai2025p845,
  title     = {Enhancing Multimodal Protein Function Prediction Through Dual-Branch Dynamic Selection with Reconstructive Pre-Training},
  author    = {Luo, Xiaoling and Chen, Peng and Liu, Chengliang and Jin, Xiaopeng and Wen, Jie and Liu, Yumeng and Wang, Junsong},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {7598--7606},
  year      = {2025},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2025/845},
  url       = {https://doi.org/10.24963/ijcai.2025/845},
}

TY  - CPAPER
TI  - Enhancing Multimodal Protein Function Prediction Through Dual-Branch Dynamic Selection with Reconstructive Pre-Training
AU  - Luo, Xiaoling
AU  - Chen, Peng
AU  - Liu, Chengliang
AU  - Jin, Xiaopeng
AU  - Wen, Jie
AU  - Liu, Yumeng
AU  - Wang, Junsong
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 7598
EP  - 7606
DO  - 10.24963/ijcai.2025/845
UR  - https://doi.org/10.24963/ijcai.2025/845
ER  -