MA‐PEP: A novel anticancer peptide prediction framework with multimodal feature fusion based on attention mechanism

• DOI: 10.1002/pro.4966
• Publication Date: 2024-03-27T07:08:05Z
• Authors: Xiao Liang; Haochen Zhao; Jianxin Wang
• Keywords: Benchmark (surveying); Feature (linguistics); Limiting

AntiCancer Peptides (ACPs) have emerged as promising therapeutic agents for cancer treatment. The time-consuming and costly nature of wet-lab discriminatory methods has spurred the development various machine learning deep learning-based ACP classification methods. Nonetheless, current encountered challenges in efficiently integrating features from peptide modalities, thereby limiting a more comprehensive understanding ACPs further restricting improvement prediction model performance. In this study, we introduce novel method, MA-PEP, which leverages multiple attention mechanisms feature enhancement fusion to improve prediction. By enhanced molecular-level chemical sequence information peptides, MA-PEP demonstrates superior performance across several benchmark datasets, highlighting its efficacy Moreover, visual analysis case studies demonstrate MA-PEP's reliable extraction capability promise realm exploration. code datasets are available at https://github.com/liangxiaodata/MA-PEP.