In silico drug discovery for COVID-19 using gene expression profiles of mouse infected by MHV and those of human cell lines infected by SARS-CoV-2

In silico drug discovery for COVID-19 using gene expression profiles of mouse infected by MHV and those of human cell lines infected by SARS-CoV-2

Y-h. Taguchi, Department of Physics, Chuo University, Tokyo 112-8551, Japan

Turki Turki, Department of Computer Science, King Abdulaziz University, Jeddah, Saudi Arabia

Developing drugs for SARS-CoV-2 is urgent, since economic activity must be suppressed if infection is tried to be suppressed without effective drugs. In spite of that, drug discovery usually takes more than ten years to complete; we cannot wait so long. Drug repositioning is a promising candidate strategy, since we can make use of a list of drug compounds that are used for other diseases. Only problem is how to select more promising candidate compounds from the list of available drugs. One possible strategy is to compare gene expression profiles of model animals and cell lines between virus infection and drug treatment. The recently proposed tensor decomposition based unsupervised feature extraction was applied to mouse liver and spleen infected by MHV that is regarded as an effective model virus of SARS-CoV as well as human cell lines infected by SARS-CoV-2. Genes associated with altered expression by infection are highly coincident with human proteins reported to be interacting with SARS-CoV-2 proteins during infection. With screening drugs that affect these genes, we identified various anti-virus drug candidate compounds. The list of selected drugs includes ivermectin that is under clinical trials.

Published Paper: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0238907