Deep Learning Algorithms Achieved Satisfactory Predictions When Trained on a Novel Collection of Anticoronavirus Molecules

Emna Harigua-Souiai; Mohamed Mahmoud Heinhane; Yosser Zina Abdelkrim; Oussama Souiai; Ines Abdeljaoued-Tej; Ikram Guizani

doi:10.3389/fgene.2021.744170

Deep Learning Algorithms Achieved Satisfactory Predictions When Trained on a Novel Collection of Anticoronavirus Molecules

Front Genet. 2021 Nov 29:12:744170. doi: 10.3389/fgene.2021.744170. eCollection 2021.

Authors

Emna Harigua-Souiai¹, Mohamed Mahmoud Heinhane¹, Yosser Zina Abdelkrim¹, Oussama Souiai², Ines Abdeljaoued-Tej^{2

3}, Ikram Guizani¹

Affiliations

¹ Laboratory of Molecular Epidemiology and Experimental Pathology-LR16IPT04, Institut Pasteur de Tunis, Université de Tunis El Manar, Tunis, Tunisia.
² Laboratory of BioInformatics BioMathematics and BioStatistics (BIMS)-LR20IPT09, Institut Pasteur de Tunis, University of Tunis El Manar, Tunis, Tunisia.
³ Engineering School of Statistics and Information Analysis, University of Carthage, Ariana, Tunisia.

Abstract

Drug discovery and repurposing against COVID-19 is a highly relevant topic with huge efforts dedicated to delivering novel therapeutics targeting SARS-CoV-2. In this context, computer-aided drug discovery is of interest in orienting the early high throughput screenings and in optimizing the hit identification rate. We herein propose a pipeline for Ligand-Based Drug Discovery (LBDD) against SARS-CoV-2. Through an extensive search of the literature and multiple steps of filtering, we integrated information on 2,610 molecules having a validated effect against SARS-CoV and/or SARS-CoV-2. The chemical structures of these molecules were encoded through multiple systems to be readily useful as input to conventional machine learning (ML) algorithms or deep learning (DL) architectures. We assessed the performances of seven ML algorithms and four DL algorithms in achieving molecule classification into two classes: active and inactive. The Random Forests (RF), Graph Convolutional Network (GCN), and Directed Acyclic Graph (DAG) models achieved the best performances. These models were further optimized through hyperparameter tuning and achieved ROC-AUC scores through cross-validation of 85, 83, and 79% for RF, GCN, and DAG models, respectively. An external validation step on the FDA-approved drugs collection revealed a superior potential of DL algorithms to achieve drug repurposing against SARS-CoV-2 based on the dataset herein presented. Namely, GCN and DAG achieved more than 50% of the true positive rate assessed on the confirmed hits of a PubChem bioassay.

Keywords: SARS-CoV-2; artificial neural network; deep learning; drug discovery and repurposing; graph convoluational networks; machine learning.