Table 3.

Machine learning and deep learning based drug repurposing studies for Covid-19.

	Study	ML Models	Dataset	Targets	Drugs	Ref.
1	AI approach fighting COVID-19 with repurposing drugs	Deep neural network	Drugs with known profiles	FIP and Covid-19	13/80 Drugs showed in-vitro activity	[80]
2	Deep Learning based prediction of Commercially available drugs	Deep learning-based, Molecule Transformer-Drug Target Interaction (MT-DTI)	SMILES and DTIs	Covid-19 associated 3CLpro, RdRp, helicase, 3′-to-5′ exonuclease, endoRNAse, and 2′-O-ribose methyltransferase	8 Drugs	[70]
3	machine learning and mechanistic models of signal transduction circuits related to SARS-CoV-2 infection	mechanistic model	Drug targets, pathway genes, Gene expression	Covid-19 related signaling pathways and transduction circuits	–	[81]
4	Open Data to Discover Therapeutics for COVID-19 Using Deep Learning	Deep learning-based knowledge graph	Knowledge graph of 15 million edges across 39 types of relationships	Covid-19	41 Drugs	[72]
5	Repositioning of 8565 Existing Drugs for COVID-19	2-D fingerprint, GBDT model, Recurrent Neural Network (RNN)	Largest available experimental data set for SARS-CoV-2 or SARS-CoV 3CL (main) protease inhibitors	Covid-19 associated 3CLpro	40 Drugs	[82]
6	Large scale virtual screening of ligand using DNN	Deep neural network enabled ChemAI	220 M data points across 3.6 M molecules from three public drug-discovery databases	Covid-19 related 3CLpro, PLP	20 Drugs	[83]
7	Deep learning enabled docking to rapidly identify the potential Covid-19 inhibitors	Deep learning-based Docking Platform	Purchasable compounds obtained from databases. (1.3 billion compounds from Zinc)	Covid-19 associated Mpro	1000 Drugs	[78]
8	Prediction of potential commercially inhibitors against SARSCoV-2 by multi-task deep model	Multi-task deep learning model for classification and regression	Covid-19 related viral protein dataset	Covid-19 related 3CLpro	10 Drugs	[84]
9	DeepPurpose: a deep learning library for drug–target interaction prediction	Convolutional neural network (CNN), Deep purpose	DTIs	Covid-19 related 3CLpro	13 Drugs	[85]
10	Screening of Therapeutic Agents for COVID-19 Using Machine Learning and Ensemble Docking Studies	Random forest (RF) regression algorithm, ensemble docking	DTIs	S, S-ACE2 complex	187 Drugs	[86]
11	SARS-CoV-2 using ML from a > 10 million chemical space	Support vector machine (SVM), Random Forest (RF)	assay data of 65 Covid-19 host targets and purchasable 14 million chemicals	Covid-19 host targets	58 Drugs	[87]