DeepAVP-TPPred: identification of antiviral peptides using transformed image-based localized descriptors and binary tree growth algorithm

• DOI: 10.1093/bioinformatics/btae305
• Publication Date: 2024-05-03T04:49:45Z
• Authors: Matee Ullah; Shahid Akbar; Ali Raza; Quan Zou
• Keywords: Identification; Tree (set theory); Random binary tree

Abstract Motivation Despite the extensive manufacturing of antiviral drugs and vaccination, viral infections continue to be a major human ailment. Antiviral peptides (AVPs) have emerged as potential candidates in pursuit novel drugs. These show vigorous activity against diverse range viruses by targeting different phases life cycle. Therefore, accurate prediction AVPs is an essential yet challenging task. Lately, many machine learning-based approaches developed for this purpose; however, their limited capabilities terms feature engineering, accuracy, generalization make these methods restricted. Results In present study, we aim develop efficient approach identification AVPs, referred DeepAVP-TPPred, address aforementioned problems. First, extract two new transformed sets using our designed image-based extraction algorithms integrate them with evolutionary information-based feature. Next, were optimized selection called binary tree growth Algorithm. Finally, optimal space from training dataset was fed deep neural network build final classification model. The proposed model DeepAVP-TPPred tested stringent 5-fold cross-validation independent testing methods, which achieved maximum performance showed enhanced efficiency over existing predictors both accuracy capabilities. Availability implementation https://github.com/MateeullahKhan/DeepAVP-TPPred.