UniBioPAN: A Novel\nUniversal Classification Architecture\nfor Bioactive Peptides Inspired by Video Action Recognition

The classification of bioactive peptides is of great\nimportance\nin protein biology, but there is still a lack of a universal and effective\nclassifier. Inspired by video action recognition, we developed the\nUniBioPAN architecture to create a universal peptide classifier to\nsolve this problem. The architecture treats the peptide sequence as\na video sequence and the molecular image of each amino acid in the\npeptide sequence as a video frame, enabling feature extraction and\nclassification using convolutional neural networks, bidirectional\nlong short-term memory networks, and fully connected networks. As\na novel peptide classification architecture, UniBioPAN significantly\noutperforms other universal architecture in ACC, AUC and MCC across\n11 data sets, and F1 score in 9 data sets. UniBioPAN is available\nin three ways: python script, jupyter notebook script and web server\n(https://gzliang.cqu.edu.cn/software/UniBioPAN.html). In summary, UniBioPAN is a universal, convenient, and high-performance\npeptide classification architecture. UniBioPAN holds significant importance\nin the discovery of bioactive peptides and the advancement of peptide\nclassifiers. All the codes and data sets are publicly available at https://github.com/sanwrh/UniBioPAN.