A New Advanced In Silico Drug Discovery Method for Novel Coronavirus (SARS-CoV-2) with Tensor Decomposition-Based Unsupervised Feature Extraction

Abstract

Background: COVID-19 is a critical pandemic that has affected human communities worldwide. Although it is urgent to rapidly develop effective drugs, large number of candidate drug compounds may be useful for treating COVID-19, and evaluation of these drugs is time-consuming and costly. Thus, screening to identify potentially effective drugs prior to experimental validation is necessary. Method: In this study, we applied the recently proposed method tensor decomposition (TD)-based unsupervised feature extraction (FE) to gene expression profiles of multiple lung cancer cell lines infected with severe acute respiratory syndrome coronavirus 2. We identified drug candidate compounds that significantly altered the expression of the 163 genes selected by TD-based unsupervised FE. Results: Numerous drugs were successfully screened, including many known antiviral drug compounds. Conclusions: The drugs screened using our strategy may be effective candidates for treating patients with COVID-19.

Drug Repositioning by Merging Active Subnetworks Validated in Cancer and COVID-19

“…We use known clinically approved drugs for benchmark diseases, and trial drugs for COVID-19 as golden standard. We show that our method has better measurable performance than C-map [Lamb et al, 2006], a widely used drug ranking portal, as well as COVID-19 specific drug rankings in [Taguchi and Turki, 2020] and [Mousavi et al, 2020]. The quality of our ranking results on COVID-19 is arguably comparable to those in [Gysi et al, 2020] and , which use however quite different strategies.…”

“…In Supplementary File S8 and figure 9 we report performance measures for the proposed COVID-19 drug repurposing rankings from Taguchi et al 2020 [Taguchi and Turki, 2020], Mousavi et al 2020 [Mousavi et al, 2020], Gysi et al 2020 [Gysi et al, 2020], and Zhou et al 2020 .…”

“…Next, we focus on some results in the lines of research that are most relevant for our study. Taguchi et al [Taguchi and Turki, 2020] apply an unsupervised method based on tensor decomposition to perform feature extraction of gene expression profiles in multiple lung cancer cell lines infected with severe acute respiratory syndrome coronavirus 2 [Blanco-Melo et al, 2020]. They thus identified, using Enrichr [Kuleshov et al, 2016], drug candidates that significantly altered the expression of the 163 genes selected in the previous phase.…”

How glycobiology can help us treat and beat the COVID-19 pandemic

“…RBD amino acids described to be involved in the interaction with Amikacin (R403, E406, K417, Y453, and Y495) are interspersed with those interacting with HS (R346, F347, S349, N354, R355, K444, G447, Y449, Y451 and R466), thus suggesting to be all in the same glycan-binding pocket. A new advanced in silico drug discovery method for novel coronavirus (SARS-CoV-2) with tensor decomposition-based unsupervised feature extraction described Gentamicin aminoglycoside antibiotic as candidate drug ( 84 ). Insights from a molecular mechanics-assisted structure-based virtual screening experiment showed Lividomycin aminoglycoside antibiotic as a potential ACE2 inhibitor in the COVID-19 pandemic ( 85 ) ( Table 1 and Figure 5 ).…”

“…The identification of potential inhibitors of SARS-CoV-2 main protease using molecular docking studies revealed flavonoid glycosides as candidate molecules ( 98 ). Flavonoid glycosides are polyphenolic structures with covalent linkage of sugars, and Hesperidin, Rutin and Quercitrin were identified as ligands of SARS-CoV-2 main protease, RNA-dependent RNA polymerase and S glycoprotein RBD ( 84 , 99 , 100 , 101 , 102 ) ( Table 1 and Figure 5 ). These repurposed FDA-approved drugs could protect against SARS-CoV-2 infection, where an important endothelial dysfunction probably associated to systemic complications is produced ( 103 ).…”