Network analysis and disease subnets for the SARS-CoV-2/Human interactome

TL;DR

This study constructs and analyzes virus-host and host-host protein interaction networks for SARS-CoV-2 to identify key proteins and disease associations, revealing insights into COVID-19 pathology and comorbidities.

Contribution

It introduces two novel interaction networks for SARS-CoV-2 and identifies key proteins and disease links, especially highlighting the role of diabetes-related subnetworks.

Findings

Identified UBC and ORF7A as key proteins in the networks.

Discovered a significant association between the virus and diabetes-related proteins.

Highlighted the importance of host-host interactions in understanding COVID-19.

Abstract

Motivation: With the aim to amplify and make sense of interactions of virus-human proteins in the case of SARS-CoV-2, we performed a structural analysis of the network of protein interactions obtained from the integration of three sources: 1) proteins of virus SARS-CoV-2, 2)physical interactions between SARS-CoV-2 and human proteins, 3) known interactions of these human proteins between them and the dossier of affections in which these proteins are implicated. Results: As a product of this research, we present two networks, one from the interactions virus-host, and the other restricted to host-host, the last one is not usually considered for network analysis. We identified the most important proteins in both networks, those that have the maximal value of calculated invariants, these proteins are considered as the most: affected, connected or those that best monitor the flow of information in the network, among them we find UBC, a human protein related with ubiquination, linked with different stages of coronavirus disease, and ORF7A a virus protein that induces apoptosis in infected cells, associated with virion tethering. Using the constructed networks, we establish the more significant diseases corresponding with human proteins and their connections with other proteins. It is relevant that the identified diseases coincide with comorbidities, particularly the subnetwork of diabetes involves a great quantity of virus and human proteins (56%) and interactions (60%), this could explain the effect of this condition as an important cause of disease complications.