A computational model for predicting fusion peptide of retroviruses
Institution:
1. School of Electronic and Information Engineering, Xi’an Jiaotong University, Xi’an, China;2. School of Electrical Engineering, Xi’an Jiaotong University, Xi’an, China;1. Department of Blood Transfusion, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China;2. Department of Gastrointestinal Surgery, The Affiliated Hospital of Qingdao University, Qingdao 266003, China;1. Robert Koch Institute, HIV and Other Retroviruses, Nordufer 20, 13353 Berlin, Germany;2. University of the Sunshine Coast, Locked Bag 4, 4558 Maroochydore DC, Queensland, Australia
Abstract:
As a pivotal domain within envelope protein, fusion peptide (FP) plays a crucial role in pathogenicity and therapeutic intervention. Taken into account the limited FP annotations in NCBI database and absence of FP prediction software, it is urgent and desirable to develop a bioinformatics tool to predict new putative FPs (np-FPs) in retroviruses. In this work, a sequence-based FP model was proposed by combining Hidden Markov Method with similarity comparison. The classification accuracies are 91.97% and 92.31% corresponding to 10-fold and leave-one-out cross-validation. After scanning sequences without FP annotations, this model discovered 53,946 np-FPs. The statistical results on FPs or np-FPs reveal that FP is a conserved and hydrophobic domain. The FP software programmed for windows environment is available at https://sourceforge.net/projects/fptool/files/?source=navbar.
