Table 2.

Application of artificial intelligence (AI) algorithms including machine learning (ML) and deep learning principles in drug design and discovery process

Tool and Software	Description	Method	Feature	Reference
QSAR modeling
QSAR-Co-X	Open-source toolkit for multi-target QSAR modeling. https://github.com/ncordeirfcup/QSAR-Co-X	Machine learning and classification model	Integrate diverse chemical and biological data into a single model equation	[239]
Cloud 3D-QSAR	A web tool for the development of quantitative structure–activity relationship models in drug discovery. http://agroda.gzu.edu.cn:9999/ccb/server/cloud3dQSAR/	Machine learning	Integrating the functions of molecular structure generation, alignment, molecular interaction field (MIF)	[244]
ChemDes	An integrated web-based platform for molecular descriptor and fingerprint computation. http://www.scbdd.com/chemdes	Pybel, CDK, RDKit, BlueDesc, Chemopy, PaDEL and jCompoundMapper	Format converting, MOPAC optimization and fingerprint similarity calculation	[379]
OntoQSAR	An Ontology for Interpreting Chemical and Biological Data in Quantitative Structure–Activity Relationship Studies	Machine learning mathematical model	Obtain chemical descriptors and biological properties of chemical compounds	[380]
ChemGrapher	Optical graph recognition of chemical compounds	Deep learning	Produces all information necessary to relate each component of the resulting graph to the source image	[381]
ChemSAR	An online pipelining platform for molecular SAR modeling. http://chemsar.scbdd.com/	RDKit or ChemoPy package, scikit-learn package	Generating SAR classification models that will benefit cheminformatics and other biomedical users	[382]
ANFIS	Evaluate physicochemical descriptors of certain chemical compounds for their appropriate biological activities in terms of QSAR models with the aid of artificial neural network (ANN) approach combined with the principle of fuzzy logic	Neuro-fuzzy modeling and principal component analysis	ANFIS was applied to train the final descriptors (Mor22m, E3s, R3v + , and R1e +) using a hybrid algorithm consisting of back-propagation and least-square estimation while the optimum number and shape of related functions were obtained through the subtractive clustering algorithm	[383]
Drug repurposing
DrugNet	Network-based drug-disease prioritization by integrating heterogeneous data. http://genome2.ugr.es/drugnet/	Machine learning	Simultaneous integration of information about diseases, drugs and targets can lead to a significant improvement in drug repositioning tasks	[274]
RepCOOL	Computational drug repositioning via integrating heterogeneous biological networks	Random forest classifier	The potency of the proposed method in detecting true drug-disease relationships	[282]
GIPAE	Computational drug repositioning, designed to identify new indications for existing drugs,	Gaussian interaction profile kernel and autoencoder	The batch normalization layer and the full-connected layer are introduced to reduce training complexity	[384]
DrPOCS	Drug Repositioning Based on Projection onto Convex Sets	Machine learning	DrPOCS predicts potential associations between drugs and diseases with matrix completion	[385]
HeteroDualNet	A dual convolutional neural network with heterogeneous layers for drug-disease association prediction via chou’s five-step rule	Neural network	Embedded heterogeneous layers of original and neighboring drug-disease representations in a dual neural network improved the association prediction performance	[386]
RCDR	A Recommender Based Method for Computational Drug Repurposing	Collaborative filtering model	Prioritize candidate drugs for diseases	[387]
GRTR	Drug-disease association prediction based on graph regularized transductive regression on a heterogeneous network	Regression model	Graph-regularized transductive regression is used to score and rank drug-disease associations iteratively	[388]
SAEROF	An ensemble approach for large-scale drug-disease association prediction by incorporating rotation forest and sparse autoencoder deep neural network	Deep neural network	This model is a feasible and effective method to predict drug-disease correlation, and its performance is significantly improved compared with existing methods	[389]
WGMFDDA	A novel weighted-based graph regularized matrix factorization for predicting drug-disease associations	K -nearest neighbor	The framework of graph regularized matrix factorization is utilized to reveal unknown associations of drugs with the disease. To evaluate the prediction performance of the proposed WGMFDDA method, ten-fold cross-validation is performed on Fdata set	[390]
HNet-DNN	Inferring new drug–disease associations with deep neural network based on heterogeneous network features	Deep neural network	Topological features for drug-disease associations from the heterogeneous network and used them to train a DNN model	[391]
DeepConv-DTI	Prediction of drug-target interactions via deep learning with convolution on protein sequences. https://github.com/GIST-CSBL/DeepConv-DTI	Deep learning	Prediction model for detecting local residue patterns of target proteins successfully enriches the protein features of a raw protein sequence, yielding better prediction results than previous approaches	[392]
DeepH-DTA	Predicting Drug-Target Interactions. https://github.com/Hawash-AI/deepH-DTA	Deep learning	Heterogeneous graph attention (HGAT) model to learn topological information of compound molecules and bidirectional ConvLSTM layers for modeling spatio-sequential information in simplified molecular-input line-entry system (SMILES) sequences of drug data	[393]
Neg Stacking	Drug-target interaction prediction. https://github.com/Open-ss/NegStacking	Ensemble learning and logistic regression	NegStacking can improve the performance of predictive DTIs, and it has broad application prospects for improving the drug discovery process	[394]
SPIDR	Small-molecule peptide-influenced drug repurposing	Genetic algorithm and heuristic search procedure	SPIDR has been generalized and integrated into DockoMatic v 2.1	[395]
DeepPurpose	Library for drug-target interaction prediction. https://github.com/kexinhuang12345/DeepPurpose	Deep learning	Supports the training of customized DTI prediction models by implementing 15 compound and protein encoders and over 50 neural architectures, along with providing many other useful features	[396]
DTI-CDF	A cascade deep forest model toward the prediction of drug-target interactions based on hybrid features. https://github.com/a96123155/DTI-CDF	Deep forest model	There are 1352 newly predicted DTIs that are proved to be correct by KEGG and DrugBank databases	[397]
Pred-binding	Large-scale protein–ligand binding affinity prediction	Support vector machine and random forest	1589 molecular descriptors and 1080 protein descriptors in 9948 ligand–protein pairs predicted DTIs that were quantified by Ki values. The cross-validation coefficient of determination of 0.6079 for SVM and 0.6267 for RF was obtained, respectively	[398]
Physicochemical properties and bioactivity
Chembench	A Publicly Accessible, Integrated Cheminformatics Portal. https://chembench.mml.unc.edu	Machine learning	Tools and services for computer-assisted drug design and computational toxicology available on Chembench	[399]
mCSM-lig	Quantifying the effects of mutations on protein-small molecule affinity in genetic disease and emergence of drug resistance. http://structure.bioc.cam.ac.uk/mcsm_lig	Machine learning models, Platinum database	Effective in predicting a range of chemotherapeutic, antiviral and antibiotic resistance mutations, providing useful insights for genotypic screening and guiding drug development	[400]
CSM-lig	A web server for assessing and comparing protein-small molecule affinities. http://structure.bioc.cam.ac.uk/csm_lig	Machine learning, graph-based chemical signatures based on PDBbind databases	Automatically predict binding affinities of collections of structures and assess the interactions made	[401]
mCSM-AB	A web server for predicting antibody-antigen affinity changes upon mutation. http://structure.bioc.cam.ac.uk/mcsm_ab	Machine learning	Predicting antibody-antigen affinity changes upon mutation which relies on graph-based signatures	[402]
dendPoint	A web resource for dendrimer pharmacokinetics investigation and prediction. http://biosig.unimelb.edu.au/dendpoint	Machine learning and principal component analysis	Used to guide dendrimer construct design and refinement before embarking on more time-consuming and expensive in vivo testing	[403]
MDCKpred	A web tool to calculate MDCK permeability coefficient of small molecule using membrane-interaction chemical features. http://www.mdckpred.in/	Regression model	An intuitive way of prioritizing small molecules based on calculated MDCK permeabilities	[404]
Vienna LiverTox	Prediction of interactions profiles of small molecules with transporters relevant for regulatory agencies. https://livertox.univie.ac.at/	Machine learning classification model	Identify pharmacokinetic properties	[405]
Ambit-SMIRKS	A software module for reaction representation, reaction search and structure transformation. http://ambit.sourceforge.net/smirks	The Chemistry Development Kit	Standardization of large chemical databases and pathway transformation database and prediction	[406]
COSMOfrag	A Novel Tool for High-Throughput ADME Property Prediction and Similarity Screening	Quantum Chemistry	In the COSMO − RS picture, any molecular information is gathered in the so-called σ profiles, COSMOfrag replaces the single σ profile with a composition of partial σ profiles, selected by the use of extensive similarity searching algorithms	[407]
RosENet	Predicting the absolute binding affinity of protein–ligand complexes	Convolutional neural networks	Combines voxelized molecular mechanics energies and molecular descriptors	[408]
MDeePred	Novel multi-channel protein featurization for deep learning-based binding affinity prediction in drug discovery. https://github.com/cansyl/MDeePred	Deep learning	MDeePred is a scalable method with sufficiently high predictive performance	[409]
Mode of action and toxicity of compounds
ProTox-II	Webserver for the prediction of toxicity of chemicals. http://tox.charite.de/protox_II	Molecular similarity, fragment propensities, and machine learning	Predicts acute toxicity, hepatotoxicity, cytotoxicity, carcinogenicity, mutagenicity, immunotoxicity	[410]
ADMETlab	A platform for systematic ADMET evaluation based on a comprehensively collected ADMET database. http://admet.scbdd.com/	Designed based on the Django framework in Python	Early drug-likeness evaluation, rapid ADMET virtual screening or filtering and prioritization of chemical structures	[411]
lazar	A modular predictive toxicology framework	QSAR model, classification model, and regression model	Choose between a large variety of algorithms for descriptor calculation and selection, chemical similarity indices, and model building	[412]
TargetNet	A web service for predicting potential drug-target interaction profiling via multi-target SAR models. http://targetnet.scbdd.com	Naïve Bayes models	The server will predict the activity of the user's molecule across 623 human proteins by the established high-quality SAR model, thus generating a DTI profiling that can be used as a feature vector of chemicals for wide applications	[413]
PSBP-SVM	The computational identifier for predicting polystyrene binding peptides. http://server.malab.cn/PSBP-SVM/index.jsp	Machine learning: support vector machines	Model contains four machine learning steps, including feature extraction, feature selection, model training and optimization	[414]
IDDkin	Prediction of kinase inhibitors. https://github.com/CS-BIO/IDDkin	Deep diffusion model	Network-based computational methods could be employed to aggregate the effective information from heterogeneous sources	[415]
SMPDB 2.0	Comprehensive, colorful, fully searchable and highly interactive database for visualizing human metabolic, drug action, drug metabolism, physiological activity and metabolic disease pathways. http://www.smpdb.ca/		Because of its utility and breadth of coverage, SMPDB is now integrated into several other databases, including HMDB and DrugBank	[416]
DruGeVar	Online resource triangulating drugs with genes and genomic biomarkers for clinical pharmacogenomics. http://drugevar.genomicmedicinealliance.org		Allows users to formulate simple and complex queries	[417]
DrugPathSeeker	Interactive UI for exploring drug-ADR relation via pathways	Machine learning	Uses a Small Molecular Risk Profiler to make ADR predictions for a given drug	[418]
SNF-NN	Computational method to predict drug-disease interactions	Neural networks	Computational drug repositioning research can significantly benefit from integrating similarity measures in heterogeneous networks	[419]
DeepDrug	A general graph-based deep learning framework for drug relation prediction. https://github.com/wanwenzeng/deepdrug	Graph convolutional networks	The structural features learned by DeepDrug, which display compatible and accordant patterns in chemical properties, providing additional evidence to support the strong predictive power of DeepDrug	[420]