Abstract
Many infectious diseases are caused by bacterial pathogens. The pathogenic mechanisms of bacterial pathogens are complex and it is usually caused by virulence factors (VFs) in many cases. Whether VFs exist is the main difference between the genomes of pathogenic and non-pathogenic bacteria. Therefore, identification of VFs is of great significance for exploring the pathogenesis of infectious diseases. In this paper, we develop a new model for predicting VFs based on multiple features and deep learning. Firstly, we used kmer, Dipeptide deviation from expected mean (DDE) and Amino acid composition (AAC) to extract features. And the deep model is constructed by integrating residual block and gated recurrent unit (GRU). Residual neural network is a kind of jump connection network to avoid gradient explosion or gradient disappearance, which is used to train deeper networks. Two residual blocks, which are separated by Relu activation function and Batch Normalization(BN) layer are used. Finally, the combined features processed by resnet are input into the recurrent structure GRU, and important information is filtered out through reset gate and update gate. The results of the proposed model are better than those of the existing methods by 10-fold cross-validation, and the accuracy is 93.81% and 95.43%, respectively. It shows that the deep hybrid model for identifying virulence factors based on residual block and gate recurrent unit is effective and feasible. The codes and datasets are accessible at https://github.com/VFs625/PreVFs-RG.git.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: IEEE/ACM Transactions on Computational Biology and Bioinformatics
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.