Abstract:
Host-pathogen interactions (HPIs) play vital roles in many biological processes and are directly involved in infectious diseases. HPIs are involved in every step of disease: from the initial pathogen transmission, through the activation of the pathogenic mechanism used to overcome (or hijack) host cell defenses, and to the pathogen becoming established and massively reproduce within the host system; all these processes involve the interaction of host and pathogen proteins. Identification of these protein–protein interactions (PPIs) could help to unravel the disease pathway, provide methods to improve resistance and ultimately accelerate the development disease control strategies. During the last decade, experimental methods to identify HPIs have been used to decipher host–pathogen systems with the caveat that those techniques are labor-intensive, expensive and time-consuming. To complement experimental methods for identifying PPIs, many computational methods have been developed to accelerate the discovery of PPIs. Since computational tools offer a promising alternative, we developed web-based R/Shiny app, Deep-HPI-pred, that uses network-driven feature learning method to predict the yet unmapped interactions between pathogen and host proteins. Using citrus and the CLas bacteria training sets as a case study, we show that Deep-HPI-pred achieves a good performance with PPI predictions between citrus and the CLasbacteria. The Deep-HPI-pred is the first to use Multilayer Perceptron (MLP) models for HPI prediction. These models have been selected based on a comprehensive evaluation of topological features and neural network architectures. The best prediction models have been tested on independent validation datasets, which achieved an overall Matthews correlation coefficient (MCC) value of greater than 0.80 for host–pathogen using the eigenvector centrality topologicalfeature. Additionally, the interaction network also offers relevant gene ontology (GO) term information for each pathogen and host protein within the system. This protein annotation data contributes an additional layer to our understanding of the intricate dynamics within host-pathogen interactions. Collectively, Deep-HPI-pred applet integrates both the detection and visualization of interaction networks in a single web service, facilitating the apprehension of model and non-model host–pathogen systems to aid the researchers in building hypotheses and designing appropriate experiments
Page(s):
75-75
DOI:
DOI not available
Published:
Journal: Abstract Book on International Conference on Food and Applied Sciences (ICFAS-23) 3-5 August 23, Volume: 0, Issue: 0, Year: 2023
Keywords:
Neural Networks
,
deep learning
,
Prediction
,
multilayer perceptron
,
Hostpathogen interactions
,
topological features