Awesome

👏 A Survey of Artificial Intelligence in Drug Discovery

💡 Artificial intelligence has been widely applied in drug discovery over the past decade and is still gaining popularity. This repository compiles a collection works on related areas, based on the manuscript Artificial Intelligence in Drug Discovery: Applications and Techniques by Jianyuan Deng et al. The preprint version is available in ResearchGate. Hope you will find it useful for your research (citation is provided below).

🔔 This repository is updated regularly.

@article{deng2022artificial,
  title={Artificial intelligence in drug discovery: applications and techniques},
  author={Deng, Jianyuan and Yang, Zhibo and Ojima, Iwao and Samaras, Dimitris and Wang, Fusheng},
  journal={Briefings in Bioinformatics},
  volume={23},
  number={1},
  pages={bbab430},
  year={2022},
  publisher={Oxford University Press}
}

Contents

• Reviews and Perspectives

  • General Drug Discovery
  • Drug Discovery in the AI Era
  • AI-Driven Drug Discovery: Hope or Hype

• Data, Representation and Benchmarks

  • Large-Scale Databases
    • PubChem
    • ChEMBL
    • ZINC
    • Others
  • Molecular Representations
  • Benchmark Platforms
    • MoleculeNet
    • MolMapNet
    • ChemProp
    • REINVENT
    • Guacamol
    • MOSES
    • GraphINVENT
    • ATOM3D

• Model Architectures

  • Convolutional Neural Networks
  • Recurrent Neural Networks
  • Graph Neural Networks
  • Variational Autoencoders
  • Generative Adversarial Networks
  • Normalizing Flow Models
  • Transformers

• Learning Paradigms

  • Self-Supervised Learning
    • Generative Learning
    • Contrastive Learning
  • Reinforcement Learning
  • Other Learning Paradigms
    • Metric Learning
    • Few-Shot Learning
    • Meta Learning
    • Active Learning

• Addressing Existing Challenges

1. Reviews and Perspectives

• Integration of virtual and high-throughput screening (Nat Rev Drug Discov 2002) [Paper]
• Chemical space and biology (Nature 2004) [Paper]
• Computer-based de novo design of drug-like molecules (Nat Rev Drug Discov 2005) [Paper]
• On Outliers and Activity Cliffs-Why QSAR Often Disappoints (J Chem Inf Model 2006) [Paper]
• Evaluating Virtual Screening Methods:  Good and Bad Metrics for the “Early Recognition” Problem (J Chem Inf Model 2007) [Paper]
• Virtual screening: an endless staircase? (Nat Rev Drug Discov 2010) [Paper]
• Privileged Scaffolds for Library Design and Drug Discovery (Curr Opin Chem Biol 2010) [Paper]
• Principles of early drug discovery (Br J Pharmacol 2011) [Paper]
• Recognizing Pitfalls in Virtual Screening: A Critical Review (J Chem Inf Model 2012) [Paper]
• Multi-objective optimization methods in drug design (Drug Discov Today 2013) [Paper]
• Finding the rules for successful drug optimisation (Drug Discov Today 2014) [Paper]
• Recent Progress in Understanding Activity Cliffs and Their Utility in Medicinal Chemistry (J Med Chem 2014) [Paper]
• Automating Drug Discovery (Nat Rev Drug Discov 2017) [Paper]
• Interpretation of Quantitative Structure−Activity Relationship Models: Past, Present, and Future (J Chem Inf Model 2017) [Paper]
• Advances and Challenges in Computational Target Prediction (J Chem Inf Model 2019) [Paper]
• Duality of activity cliffs in drug discovery (Expert Opin Drug Discov 2019) [Paper]
• QSAR without borders (Chem Soc Rev 2020) [Paper]
• Designing small molecules for therapeutic success: A contemporary perspective (Drug Discov Today 2021) [Paper]
• Phenotypic drug discovery: recent successes, lessons learned and new directions (Nat Rev Drug Discov 2022) [Paper]
• Is the reductionist paradox an Achilles Heel of drug discovery? (J Comput Aided Mol 2022) [Paper]

• Machine-learning approaches in drug discovery: methods and applications (Drug Discov Today 2015) [Paper]
• The rise of deep learning in drug discovery (Drug Discov Today 2018) [Paper]
• Applications of machine learning in drug discovery and development (Nat Rev Drug Discov 2019)[Paper]
• Deep Learning in Chemistry (J Chem Inf Model 2019) [Paper]
• Deep learning for molecular design—a review of the state of the art (Mol Syst Des Eng 2019) [Paper]
• Efficient molecular encoders for virtual screening (Drug Discov Today Technol 2019) [Paper]
• Artificial intelligence in chemistry and drug design (J Comput Aid Mol Des 2020) [Paper]
• Graph convolutional networks for computational drug development and discovery (Brief Bioinformatics 2020) [Paper]
• Transfer Learning for Drug Discovery (J Med Chem 2020) [Paper]
• Learning Molecular Representations for Medicinal Chemistry (J Med Chem 2020) [Paper]
• Exploring chemical space using natural language processing methodologies for drug discovery (Drug Discov Today 2020) [Paper]
• Practical Notes on Building Molecular Graph Generative Models (Applied AI Letters 2020) [Paper]
• A compact review of molecular property prediction with graph neural networks (Drug Discov Today 2020) [Paper]
• Artificial intelligence in drug discovery: Recent advances and future perspectives (Expert Opin Drug Discov 2021) [Paper]
• Artificial intelligence in drug discovery and development (Drug Discov Today 2021) [Paper]
• Graph neural networks for automated de novo drug design (Drug Discov Today 2021) [Paper]
• De novo molecular design and generative models (Drug Discov Today 2021) [Paper]
• Artificial Intelligence for Drug Discovery (KDD 2021) [Paper] [Website] [TorchDrug]
• Generative Deep Learning for Targeted Compound Design (J Chem Inf Model 2021) [Paper]
• Explainable Machine Learning for Property Predictions in Compound Optimization (J Med Chem 2021) [Paper]
• A decade of machine learning-based predictive models for human pharmacokinetics: Advances and challenges (Drug Discov Today 2021) [Paper]
• Defining Levels of Automated Chemical Design (J Med Chem 2022) [Paper]
• Evaluation guidelines for machine learning tools in the chemical sciences (Nat Rev Chem 2022) [Paper]
• Combining DELs and machine learning for toxicology prediction (Drug Discov Today 2022) [Paper]

<i>Side Notes: Successful Applications </i>

• Deep learning enables rapid identification of potent DDR1 kinase inhibitors (aka: GENTRL; Nat Biotechnol 2019) [Paper] [Code] (Insilico Medicine)
• A Deep Learning Approach to Antibiotic Discovery (Cell 2020) [Paper] [Code] (MIT CSAIL)
• "BenevolentAI Announces First Patient Dosed In Its Atopic Dermatitis Clinical Trial" [Link] (BenevolentAI)
• "Exscientia Announces First AI-Designed Immuno-Oncology Drug to Enter Clinical Trials" [Link] (Exscientia)
• "Breaking Big Pharma's AI barrier: Insilico Medicine uncovers novel target, new drug for pulmonary fibrosis in 18 months" [Link] (Insilico Medicine)

• Rethinking drug design in the artificial intelligence era (Nat Rev Drug Discov 2020) [Paper]
• Towards reproducible computational drug discovery (J Cheminf 2020) [Paper]
• Current Trends, Overlooked Issues, and Unmet Challenges in Virtual Screening (J Chem Inf Model 2020) [Paper]
• Drug discovery with explainable artificial intelligence (Nat Mach Intell 2020) [Paper]
• Artificial intelligence in drug discovery: what is realistic, what are illusions? Part 1: Ways to make an impact, and why we are not there yet (Drug Discov Today 2021) [Paper]
• Artificial intelligence in drug discovery: what is realistic, what are illusions? Part 2: a discussion of chemical and biological data used for AI in drug discovery (Drug Discov Today 2021) [Paper]
• Critical assessment of AI in drug discovery (Expert Opin Drug Discov 2021) [Paper]
• An Insight into Artificial Intelligence in Drug Discovery: An Interview with Professor Gisbert Schneider (Expert Opin Drug Discov 2021) [Paper]

2. Data, Representation & Benchmarks

<b>2.1 Large-Scale Databases</b>

• PubChem in 2021: new data content and improved web interfaces (Nucleic Acids Res 2021) [Paper] [Website] [Download]

• The ChEMBL database in 2017 (Nucleic Acids Res 2017) [Paper] [Website] [Download] [WebAPI]

• ZINC 15 – Ligand Discovery for Everyone (J Chem Inf Model 2015) [Paper] [Website]

• Recent applications of deep learning and machine intelligence on in silico drug discovery: methods, tools and databases (Brief Bioinformatics 2019) [Paper]
• <b>DrugBank</b> --- DrugBank 5.0: a major update to the DrugBank database for 2018 (Nucleic Acids Res 2018) [Paper] [Website] [Download]
• <b>KEGG</b> --- KEGG as a reference resource for gene and protein annotation (Nucleic Acids Res 2016) [Paper] [Website] [Download]
• <b>PDBbind</b> --- PDB-wide collection of binding data: current status of the PDBbind database (Bioinformatics 2015) [Paper] [Website] [Download]
• <b>BindingDB</b> --- BindingDB in 2015: A public database for medicinal chemistry, computational chemistry and systems pharmacology (Nucleic Acids Res 2016) [Paper] [Website] [Download]
• <b>DUD</b> --- Benchmarking Sets for Molecular Docking (J Med Chem 2006) [Paper] [Website]
• <b>DUD-E</b> --- Directory of Useful Decoys, Enhanced (DUD-E): Better Ligands and Decoys for Better Benchmarking (J Med Chem 2012) [Paper] [Website]
• <b>MUV</b> --- Maximum Unbiased Validation (MUV) Data Sets for Virtual Screening Based on PubChem Bioactivity Data (J Chem Inf Model 2009) [Paper] [Website]
• <b>STITCH</b> --- STITCH: interaction networks of chemicals and proteins (Nucleic Acids Res 2008) [Paper] [Website]
• <b>GLL&GDD</b> --- Ligand and Decoy Sets for Docking to G Protein-Coupled Receptors (J Chem Inf Model 2012) [Paper] [Website]
• <b>NRLiSt BDB</b> --- NRLiSt BDB, the Manually Curated Nuclear Receptors Ligands and Structures Benchmarking Database (J Med Chem 2014) [Paper] [Website]
• <b>SIDER</b> --- The SIDER database of drugs and side effects (Nucleic Acids Res 2016) [Paper] [Website]
• <b>Offsides&Twosides</b> --- Data-driven prediction of drug effects and interactions (Sci Transl Med 2012) [Paper] [Website]
• <b>DILIrank</b> --- DILIrank: the largest reference drug list ranked by the risk for developing drug-induced liver injury in humans (Drug Discov Today 2016) [Paper] [Website]
• <b>UniProt</b> --- UniProt: the universal protein knowledgebase in 2021 (Nucleic Acids Res 2021) [Paper] [Website]
• <b>PDB</b> --- The Protein Data Bank (Nucleic Acids Res 2000) [Paper] [Website]

<b>2.2 Small Molecule Representations</b>

• Molecular representations in AI‑driven drug discovery: a review and practical guide (J Cheminf 2020) [Paper]

<b>2.3 Benchmark Platforms</b>

• MoleculeNet: a benchmark for molecular machine learning (Chem Sci 2018) [Paper] [Code] [Download]

• Out-of-the-box deep learning prediction of pharmaceutical properties by broadly learned knowledge-based molecular representations (Nat Mach Intell 2021) [Paper] [Code]

• Analyzing Learned Molecular Representations for Property Prediction (J Chem Inf Model 2019) [Paper] [Code] [Website]

• Molecular De Novo design using Recurrent Neural Networks and Reinforcement Learning (J Cheminf 2017) [Paper] [Code]
• REINVENT 2.0 – an AI Tool for De Novo Drug Design (J Chem Inf Model 2020) [Paper] [Code]

• Graph Networks for Molecular Design (aka: GraphINVENT; Mach Learn: Sci Technol 2021) [Paper] [Code]
• Practical Notes on Building Molecular Graph Generative Models (Applied AI Letters 2020) [Paper] [Code]

• GuacaMol: Benchmarking Models for de Novo Molecular Design (J Chem Inf Model 2019) [Paper] [Code]

• Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models (Front Pharmacol 2020) [Paper] [Code]

• ATOM3D: Tasks On Molecules in Three Dimensions (NeurIPS 2021) [Paper] [Code] [Website]

3. Model Architectures

<b>3.1 Convolutional Neural Networks</b>

<i>Task*: Molecular Property Prediction; Representation*: Images</i>

• Convolutional Embedding of Attributed Molecular Graphs for Physical Property Prediction (J Chem Inf Model 2017) [Paper] (Techs - CNN + SVM)
• Chemception: a deep neural network with minimal chemistry knowledge matches the performance of expert-developed QSAR/QSPR models (aka: Chemception; arXiv 2017) [Paper]
• Toxic Colors: The Use of Deep Learning for Predicting Toxicity of Compounds Merely from Their Graphic Images (aka: Toxic Colors; J Chem Inf Model 2018) [Paper]
• KekuleScope: prediction of cancer cell line sensitivity and compound potency using convolutional neural networks trained on compound images (J Cheminf 2019) [Paper] [Code] (Techs - CNN)
• Learning Drug Functions from Chemical Structures with Convolutional Neural Networks and Random Forests (J Chem Inf Model 2019) [Paper] (Techs - CNN)
• DEEPScreen: high performance drug–target interaction prediction with convolutional neural networks using 2-D structural compound representation (Chem Sci 2020) [Paper] [Code]

<i>Task*: Molecular Property Prediction; Representation*: Fingerprints</i>

• Massively Multitask Networks for Drug Discovery (arXiv 2015) [Paper]
• Convolutional Networks on Graphs for Learning Molecular Fingerprints (NeurIPS 2015) [Paper] [Code]

<i>Side Note: Molecular Structure Extraction and Recognition</i>

• Molecular Structure Extraction from Documents Using Deep Learning (J Chem Inf Model 2019) [Paper]
• DECIMER-Segmentation: Automated extraction of chemical structure depictions from scientific literature (J Cheminf 2021) [Paper]
• DECIMER: towards deep learning for chemical image recognition (J Cheminf 2020) [Paper] [Code]
• DECIMER 1.0: Deep Learning for Chemical Image Recognition using Transformers (chemRxiv 2021) [Paper]
• Img2Mol - Accurate SMILES Recognition from Molecular Graphical Depictions (Chem Sci 2021) [Paper] [Code]

<b>3.2 Recurrent Neural Networks</b>

<i>Task*: Molecular Property Prediction; Representation*: SMILES Strings</i>

• SMILES2Vec: An Interpretable General-Purpose Deep Neural Network for Predicting Chemical Properties (aka: SMILES2Vec; arXiv 2017) [Paper]
• Large-scale comparison of machine learning methods for drug target prediction on ChEMBL (aka:SmilesLSTM; Chem Sci 2018) [Paper] [Code] (Techs - RNN + GNN + Multi-Task Learning)

<i>Task*: Molecule Generation; Representation*: SMILES Strings</i>

• Molecular de‑novo design through deep reinforcement learning (aka: REINVENT; J Cheminf 2017) [Paper] (Techs - RNN: GRU + RL: Policy-gradient REINFORCE)
• Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks (aka: CharRNN; ACS Cent Sci 2018) [Paper] (Techs - Transfer Learning)
• Exploring Deep Recurrent Models with Reinforcement Learning for Molecule Design (ICLR 2018 Workshop) [Paper] (Techs - RL: Hybrid A2C, Policy-gradient PPO)
• Deep Reinforcement Learning for de novo Drug Design（aka: ReLeaSE; Sci Adv 2018）[Paper] [Code] (Techs - RL: Policy-gradient REINFORCE)
• Deep Reinforcement Learning for Multiparameter Optimization in de novo Drug Design (J Chem Inf Model 2019) [Paper] [Code] (Techs - RNN: BiLSTM + RL: Hybrid Actor-Critic)
• Scaffold-Constrained Molecular Generation (J Chem Inf Model 2020) [Paper] (Techs - RL: Policy-based Hill Climbing)

<i>Task*: Molecule Generation; Representation*: Molecular Graphs</i>

• GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models (aka: GraphRNN; ICML 2018) [Paper] [Code]
• Learning Deep Generative Models of Graphs (ICML 2018) [Paper] [Code] (Techs - RNN: LSTM)
• MolecularRNN: Generating realistic molecular graphs with optimized properties (arXiv 2019) [Paper]
• A Deep-Learning View of Chemical Space Designed to Facilitate Drug Discovery (aka: DESMILES; J Chem Inf Model 2020) [Paper]

<b>3.3 Graph Neural Networks</b>

<i>Task*: Molecular Property Prediction; Representation*: Molecular Graphs</i>

• Molecular Graph Convolutions: Moving Beyond Fingerprints (aka: Weave; J Comput Aided Mol Des 2016) [Paper]
• Convolutional Embedding of Attributed Molecular Graphs for Physical Property Prediction (J Chem Inf Model 2017) [Paper]
• Semi-supervised classification with graph convolutional networks (aka: GraphConv; ICLR 2017) [Paper] [Code]
• Neural Message Passing for Quantum Chemistry (aka: MPNN; ICML 2017) [Paper] [Code]
• SchNet: A continuous-filter convolutional neural network for modeling quantum interactions (aka: SchNet; NeurIPS 2017)[Paper] [Code]
• Low Data Drug Discovery with One-Shot Learning (ACS Cent Sci 2017) [Paper] (Techs - LSTM: BiLSTM, attLSTM + GNN + Few-Shot Learning)
• Large-scale comparison of machine learning methods for drug target prediction on ChEMBL (aka:SmilesLSTM; Chem Sci 2018) [Paper] [Code] (Techs - RNN + GNN + Multi-Task Learning)
• PotentialNet for Molecular Property Prediction (aka: PotentialNet; ACS Cent Sci 2018) [Paper] (Techs - GNN: GCNN + Multi-Task Learning)
• Molecular Property Prediction: A Multilevel Quantum Interactions Modeling Perspective (aka: MGCN; AAAI 2019) [Paper]
• Deep Learning-Based Prediction of Drug-Induced Cardiotoxicity (J Chem Inf Model 2019) [Paper] [Code] (Techs - GCN + Multi-task Learning)
• DeepChemStable: Chemical Stability Prediction with an Attention-Based Graph Convolution Network (J Chem Inf Model 2019) [Paper] (Techs - GCN + Attention)
• Analyzing Learned Molecular Representations for Property Prediction (aka: Chemrop, D-MPNN; J Chem Inf Model 2019) [Paper] [Code]
• Molecule Property Prediction Based on Spatial Graph Embedding (aka: C-SGEN; J Chem Inf Model 2019) [Paper] [Code]
• Pushing the Boundaries of Molecular Representation for Drug Discovery with the Graph Attention Mechanism (aka: Attentive FP; J Med Chem 2019) [Paper] [Code]
• Graph convolutional neural networks as” general-purpose” property predictors: the universality and limits of applicability (J Chem Inf Model 2020) [Paper]
• N-Gram Graph: Simple Unsupervised Representation for Graphs, with Applications to Molecules (aka: N-Gram Graph; NeurIPS 2019) [Paper]
• Building attention and edge message passing neural networks for bioactivity and physical–chemical property prediction (J Cheminf 2020) [Paper] [Code] (Techs - MPNN + Multi-Task Learning)
• A self‑attention based message passing neural network for predicting molecular lipophilicity and aqueous solubility (J Cheminf 2020) [Paper] [Code] (Techs - MPNN + Self-Attention: Interpretability)
• Chemically Interpretable Graph Interaction Network for Prediction of Pharmacokinetic Properties of Drug-Like Molecules (aka: CIGIN; AAAI 2020) [Paper] [Code]
• Strategies for Pre-training Graph Neural Networks (ICLR 2020) [Paper] [Code] (Techs - Self-Supervised Learning)
• Directional Message Passing for Molecular Graphs (aka: DimeNet; ICLR 2020) [Paper] [Code]
• Drug–target affinity prediction using graph neural network and contact maps (RSC Advances 2020) [Paper] (Techs - GCN + GAT)
• ASGN: An Active Semi-supervised Graph Neural Network for Molecular Property Prediction (aka: ASGN; KDD 2020) [Paper] [Code] (Techs - Active Learning)
• Meta-Learning GNN Initializations for Low-Resource Molecular Property Prediction (ICML 2020 Workshop) [Paper] [Code] (Techs - GGNN + Meta Learning: MAML, FO-MAML, ANIL)

<i>Task*: Molecule Generation; Representation*: Molecular Graphs</i>

• Multi‑objective de novo drug design with conditional graph generative model (J Cheminf 2018) [Paper] [Code] (Techs - Conditional Graph Generative Model: MolMP, MolRNN)
• Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation (aka: GCPN; NeurIPS 2018) [Paper] [Code] (Techs - GCN + RL: PPO)
• Optimization of Molecules via Deep Reinforcement Learning (aka: MolDQN; Sci Rep 2019) [Paper] (Techs - RL: Q-learning)
• Improving Molecular Design by Stochastic Iterative Target Augmentation (ICML 2020) [Paper] [Code] (Techs - VSeq2Seq/HierGNN + Semi-Supervised Learnning)
• DeepGraphMolGen, a multi‑objective, computational strategy for generating molecules with desirable properties: a graph convolution and reinforcement learning approach (J Cheminf 2020) [Paper] (Techs - GCN + RL: PPO)
• Reinforced Molecular Optimization with Neighborhood-Controlled Grammars (aka: MNCE-RL; NeurIPS 2020) [Paper] [Code] (Techs - RL: PPO)
• Graph Networks for Molecular Design (aka: GraphINVENT; Mach Learn: Sci Technol 2021) [Paper] [Code]
• De novo drug design using reinforcement learning with graph-based deep generative models (aka: RL-GraphINVENT; ChemRxiv 2021) [Paper] [Code]

<i>Side Note: Common GNN Models</i>

• <b>Recurrent GNNs</b> Gated graph sequence neural networks (aka: GGNN; ICLR 2016) [Paper] [Code]
• <b>Convolutional GNNs (Spectral-based)</b> Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering (aka: ChebNet; NeurIPS 2016) [Paper] [Code]
• <b>Convolutional GNNs (Spectral-based)</b> Semi-supervised classification with graph convolutional networks (aka: GraphConv; ICLR 2017) [Paper] [Code]
• <b>Convolutional GNNs (Spatial-based)</b> Neural message passing for quantum chemistry (aka: MPNN; ICML 2017) [Paper] [Code]
• <b>Convolutional GNNs (Spatial-based)</b> Inductive Representation Learning on Large Graphs (aka: GraphSAGE; NeurIPS 2017) [Paper] [Code]
• <b>Convolutional GNNs (Spatial-based)</b> Graph Attention Networks (aka: GAT; ICLR 2018) [Paper] [Code]
• <b>Convolutional GNNs (Spatial-based)</b> How powerful are graph neural networks? (aka: GIN; ICLR 2019) [Paper] [Code]

<b>3.4 Variational Autoencoders</b>

<i>Task*: Molecule Generation; Representation*: SMILES Strings</i>

• Automatic chemical design using a data-driven continuous representation of molecules (arXiv 2016; ACS Cent Sci 2018) [Paper] [Code] (Techs - VAE)
• Grammar Variational Autoencoder (aka: GrammarVAE; ICML 2017) [Paper]
• Application of Generative Autoencoder in De Novo Molecular Design (Mol Inform 2017) [Paper]
• Syntax-Directed Variational Autoencoder for Structured Data (aka: SD-VAE; ICLR 2018) [Paper] [Code]
• Conditional Molecular Design with Deep Generative Models （aka: Continuous SSVAE; J Chem Inf Model 2018）[Paper] [Code]
• Molecular generative model based on conditional variational autoencoder for de novo molecular design (aka: CVAE; J Cheminf 2018) [Paper] [Code] (Techs - VAE)
• Constrained Graph Variational Autoencoders for Molecule Design (aka: CGVAE; NeurIPS 2018) [Paper] [Code]
• NEVAE: A Deep Generative Model for Molecular Graphs (aka: NeVAE; AAAI 2019) [Paper] [Code]
• De Novo Molecular Design by Combining Deep Autoencoder Recurrent Neural Networks with Generative Topographic Mapping (aka: GTMVAE; J Chem Inf Model 2019) [Paper] (Techs - Autoencoder + RNN)
• Re-balancing Variational Autoencoder Loss for Molecule Sequence Generation (aka: re-balanced VAE; ACM BCB 2020) [Paper] [Code] (Techs - RNN: BiGRU + VAE)
• CogMol: Target-Specific and Selective Drug Design for COVID-19 Using Deep Generative Models (aka: CogMol; NeurIPS 2020) [Paper] [Code]

<i>VAE Variant: AAE</i>

• Application of Generative Autoencoder in De Novo Molecular Design (Mol Inform 2017) [Paper]
• druGAN: An Advanced Generative Adversarial Autoencoder Model for de Novo Generation of New Molecules with Desired Molecular Properties in Silico (aka: druGAN; Mol Pharm 2017) [Paper]
• Entangled Conditional Adversarial Autoencoder for de Novo Drug Discovery (aka: SAAE; Mol Pharm 2018) [Paper]

<i>Task*: Molecule Generation; Representation*: Molecular Graphs</i>

• GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders (aka: GraphVAE; arXiv 2018) [Paper]
• Junction Tree Variational Autoencoder for Molecular Graph Generation (aka: JT-VAE; ICML 2018) [Paper] [Code]
• Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders (aka:Regularized VAE; NeurIPS 2018)[Paper]
• Molecular Hypergraph Grammar with Its Application to Molecular Optimization (aka: MHG-VAE; ICML 2019) [Paper] [Code]
• Efficient learning of non‑autoregressive graph variational autoencoders for molecular graph generation (J Cheminf 2019) [Paper] [Code] (Techs - Non-autoregressive VAE + RL)
• Deep learning enables rapid identification of potent DDR1 kinase inhibitors (aka: GENTRL; Nat Biotechnol 2019) [Paper] [Code] (Techs - VAE + RL: REINFORCE)
• Scaffold-based molecular design using graph generative model (aka: ScaffoldVAE; arXiv 2019) [Paper]
• Learning Multimodal Graph-to-Graph Translation for Molecule Optimization (aka: VJTNN; ICLR 2019) [Paper] [Code]
• CORE: Automatic Molecule Optimization Using Copy & Refine Strategy (AAAI 2020) [Paper] [Code]
• Hierarchical Generation of Molecular Graphs using Structural Motifs (aka: HierVAE; ICML 2020) [Paper] [Code] (Techs - Hierarchical VAE)
• Compressed graph representation for scalable molecular graph generation (J Cheminf 2020) [Paper] [Code] (Techs - Non-autoregressive VAE)

<i>Side Note: Reaction & Retrosynthesis Prediction; Representation*: Molecular Graphs</i>

• Generating Molecules via Chemical Reactions (ICLR 2019 Workshop) [Paper]
• Barking up the right tree: an approach to search over molecule synthesis DAG (NeurIPS 2020) [Paper] [Code]

<b>3.5 Generative Adversarial Networks</b>

<i>Task*: Molecule Generation; Representation*: SMILES Strings</i>

• Objective-Reinforced Generative Adversarial Networks (ORGAN) for Sequence Generation Models (aka: ORGAN; ArXiv 2017) [Paper] [Code] (Techs - GAN: G-RNN, D-CNN + RL: REINFORCE)
• Optimizing distributions over molecular space. An objective-reinforced generative adversarial network for inverse-design chemistry (aka: ORGANIC; ChemRxiv 2017) [Paper] [Code] (Techs - GAN + RL: REINFORCE)
• Reinforced Adversarial Neural Computer for de Novo Molecular Design (aka: RANC; J Chem Inf Model 2018) [Paper] (Techs - GAN + RL)

<i>Task*: Molecule Generation; Representation*: Molecular Graphs</i>

• MolGAN: An implicit generative model for small molecular graphs (aka: MolGAN; ICML 2018 Workshop) [Paper] [Code-Tensorflow] [Code-PyTorch] (Techs - GAN + RL: DDPG)

<b>3.6 Normalizing Flow Models</b>

<i>Task*: Molecule Generation; Representation*: Molecular Graphs</i>

• GraphNVP: An Invertible Flow Model for Generating Molecular Graphs (aka: GraphNVP; arXiv 2019) [Paper] [Code]
• Graph Residual Flow for Molecular Graph Generation (aka: GRF; arXiv 2019) [Paper]
• GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation (aka: GraphAF; ICLR 2020) [Paper] [Code] (Techs - Flow + RL: PPO)
• MoFlow: An Invertible Flow Model for Generating Molecular Graphs (aka: MoFlow; KDD 2020) [Paper] [Code]
• GraphDF: A Discrete Flow Model for Molecular Graph Generation (aka: GraphDF; ICML 2021) [Paper]
• Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation (NeurIPS 2021) [Paper]

<b>3.7 Transformers</b>

<i>Task*: Molecular Property Prediction; Representation*: SMILES Strings</i>

• SMILES-BERT: Large Scale Unsupervised Pre-Training for Molecular Property Prediction (aka: SMILES-BERT; ACM BCB 2019) [Paper] (Techs - BERT)
• SMILES Transformer: Pre-trained Molecular Fingerprint for Low Data Drug Discovery (aka: SMILES Transformer; arXiv 2019) [Paper] [Code]
• ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Prediction (aka: ChemBERTa; arXiv 2020) [Paper] [Code]
• Molecular representation learning with language models and domain-relevant auxiliary tasks (aka: MolBERT; NeurIPS 2020 Workshop) [Paper] [Code] (Techs - BERT + Self-Supervised Learning)
• Algebraic graph-assisted bidirectional transformers for molecular property prediction (aka: AGBT; Nat Commun 2021) [Paper] [Code]
• ChemBERTa-2: Towards Chemical Foundation Models (aka: ChemBERTa-2; arXiv 2022) [Paper]

<i>Task*: Molecular Property Prediction; Representation*: Molecular Graphs</i>

• Self-Supervised Graph Transformer on Large-Scale Molecular Data (aka: GROVER; NeurIPS 2020) [Paper] [Code] (Techs - Graph Transformer + Self-Supervised Learning)

<i>Task*: Molecule Generation; Representation*: SMILES Strings</i>

• Transformer-Based Generative Model Accelerating the Development of Novel BRAF Inhibitors (ACS Omega 2021) [Paper]
• MolGPT: Molecular Generation Using a Transformer-Decoder Model (aka: MolGPT; J Chem Inf Model 2022) [Paper] [Code]

<i>Task*: Molecule Generation; Representation*: Molecular Graphs</i>

• A Model to Search for Synthesizable Molecules (aka: Molecule Chef; NeurIPS 2019) [Paper] [Code]
• Transformer neural network for protein-specific de novo drug generation as a machine translation problem (Sci Rep 2021) [Paper]

4. Learning Paradigms

<b>4.1 Self-Supervised Learning in Molecular Property Prediction</b>

<i>Generative Learning</i>

• Strategies for Pre-training Graph Neural Networks (ICLR 2020) [Paper] [Code] (Techs - Self-Supervised Learning)
• Molecular representation learning with language models and domain-relevant auxiliary tasks (aka: MolBERT; NeurIPS 2020 Workshop) [Paper] [Code] (Techs - BERT + Self-Supervised Learning)
• Self-Supervised Graph Transformer on Large-Scale Molecular Data (aka: GROVER; NeurIPS 2020) [Paper] [Code] (Techs - Graph Transformer + Self-Supervised Learning)

<i>Contrastive Learning</i>

• MolCLR: Molecular Contrastive Learning of Representations via Graph Neural Networks (ArXiv 2021) [Paper] [Code]
• Improving Molecular Contrastive Learning via Faulty Negative Mitigation and Decomposed Fragment Contrast (J Chem Inf Model 2022) [Paper] [Code]

<b>4.2 Reinforcement Learning in Molecule Generation</b>

• Objective-Reinforced Generative Adversarial Networks (ORGAN) for Sequence Generation Models (aka: ORGAN; ArXiv 2017) [Paper] [Code] (Techs - GAN: G-RNN, D-CNN + RL: Policy-gradient REINFORCE)
• Molecular de‑novo design through deep reinforcement learning (aka: REINVENT; J Cheminf 2017) [Paper] (Techs - RNN: GRU + RL: Policy-gradient REINFORCE)
• Optimizing distributions over molecular space. An objective-reinforced generative adversarial network for inverse-design chemistry (aka: ORGANIC; ChemRxiv 2017) [Paper] [Code] (Techs - GAN + RL: Policy-gradient REINFORCE)
• Reinforced Adversarial Neural Computer for de Novo Molecular Design (aka: RANC; J Chem Inf Model 2018) [Paper] (Techs - GAN + RL)
• Exploring Deep Recurrent Models with Reinforcement Learning for Molecule Design (ICLR 2018 Workshop) [Paper] (Techs - RL: Hybrid A2C, Policy-gradient PPO)
• MolGAN: An implicit generative model for small molecular graphs (aka: MolGAN; ICML 2018 Workshop) [Paper] [Code-Tensorflow] [Code-PyTorch] (Techs - GAN + RL: Hybrid Actor-Critic DDPG)
• Deep Reinforcement Learning for de novo Drug Design（aka: ReLeaSE; Sci Adv 2018）[Paper] [Code] (Techs - RL: Policy-gradient REINFORCE)
• Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation (aka: GCPN; NeurIPS 2018)[Paper] [Code] (Techs - GCN + RL: Policy-gradient PPO)
• Deep learning enables rapid identification of potent DDR1 kinase inhibitors (aka: GENTRL; Nat Biotechnol 2019) [Paper] [Code] (Techs - VAE + RL: Policy-gradient REINFORCE)
• Deep Reinforcement Learning for Multiparameter Optimization in de novo Drug Design (aka: DeepFMPO; J Chem Inf Model 2019) [Paper] [Code] (Techs - RNN: BiLSTM + RL: Hybrid Actor-Critic)
• Optimization of Molecules via Deep Reinforcement Learning (aka: MolDQN; Sci Rep 2019) [Paper] (Techs - RL: Value-based Double Q-learning)
• Efficient learning of non‑autoregressive graph variational autoencoders for molecular graph generation (J Cheminf 2019) [Paper] [Code] (Techs - Non-autoregressive VAE + RL: Policy-gradient)
• GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation (aka: GraphAF; ICLR 2020) [Paper] [Code] (Techs - Flow + RL: Policy-gradient PPO)
• Reinforcement Learning for Molecular Design Guided by Quantum Mechanics (cka: MolGym; ICML 2020) [Paper] (Techs - RL: Policy-gradient PPO)
• DeepGraphMolGen, a multi‑objective, computational strategy for generating molecules with desirable properties: a graph convolution and reinforcement learning approach (aka: DeepGraphMolGen; J Cheminf 2020) [Paper] (Techs - GCN + RL: Policy-gradient PPO)
• Reinforced Molecular Optimization with Neighborhood-Controlled Grammars (aka: MNCE-RL; NeurIPS 2020) [Paper] [Code] (Techs - RL: Policy-gradient PPO)
• Deep inverse reinforcement learning for structural evolution of small molecules (Brief Bioinform 2021) [Paper] [Code] (Techs - Inverse RL)

<i>Side Note: Common RL Algorithms</i>

• <b>Value-based</b> Playing Atari with Deep Reinforcement Learning (aka: DQN; NeurIPS Workshop 2013) [Paper]
• <b>Value-based</b> Human-level control through deep reinforcement learning (aka: DQN; Nature 2015) [Paper]
• <b>Value-based</b> Deep Reinforcement Learning with Double Q-learning (aka: Double Q-learning; AAAI 2016) [Paper]
• <b>Value-based</b> Prioritized Experience Replay (aka: DQN with Experience Replay; ICLR 2016) [Paper]
• <b>Value-based</b> Dueling Network Architectures for Deep Reinforcement Learning (aka: Dueling Network; ICML 2016) [Paper]
• <b>Policy-gradient</b> Simple statistical gradient-following algorithms for connectionist reinforcement learning (aka: REINFORCE; Mach Learn 1992) [Paper]
• <b>Policy-gradient</b> Policy Gradient Methods for Reinforcement Learning with Function Approximation (aka: Random Policy Gradient; NeurIPS 1999) [Paper]
• <b>Policy-gradient</b> Deterministic Policy Gradient Algorithms (aka: DPG; ICML 2014) [Paper]
• <b>Policy-gradient</b> Trust Region Policy Optimization (aka: TRPO; ICML 2015) [Paper]
• <b>Policy-gradient</b> Proximal Policy Optimization Algorithms (aka: PPO; arXiv 2017 2015) [Paper]
• <b>Hybrid</b> Continuous control with deep reinforcement learning (aka: DDPG; ICLR 2016) [Paper]
• <b>Hybrid</b> Asynchronous Methods for Deep Reinforcement Learning (aka: A3C; ICML 2016) [Paper]

<i>Side Note: Pareto Optimality</i>

• De Novo Drug Design of Targeted Chemical Libraries Based on Artificial Intelligence and Pair-Based Multiobjective Optimization (J Chem Inf Model 2020) [Paper] [Code] (Techs - Pareto Optimality)
• Multiobjective de novo drug design with recurrent neural networks and nondominated sorting (J Cheminf 2020) [Paper] (Techs - Pareto Optimality)
• DrugEx v2: De Novo Design of Drug Molecule by Pareto-based Multi-Objective Reinforcement Learning in Polypharmacology (ChemRxiv) [Paper] (Techs - Pareto Optimality)

<i>Side Note: Reaction & Retrosynthesis Optimization</i>

• Optimizing chemical reactions with deep reinforcement learning (ACS Cent Sci 2017) [Paper]

<b>4.4 Other Learning Paradigms</b>

<i>Metric Learning</i>

• Machine-guided representation for accurate graph-based molecular machine learning (Phys Chem Chem Phys 2020) [Paper]
• Embedding of Molecular Structure Using Molecular Hypergraph Variational Autoencoder with Metric Learning (Mol Inform 2020) [Paper]

<i>Few-Shot Learning</i>

• Low Data Drug Discovery with One-Shot Learning (ACS Cent Sci 2017) [Paper] (Techs - LSTM: BiLSTM, attLSTM + GNN + Few-Shot Learning)
• Few-Shot Graph Learning for Molecular Property Prediction (WWW 2021) [Paper] [Code]

<i>Meta Learning</i>

• Meta-Learning GNN Initializations for Low-Resource Molecular Property Prediction (ICML 2020 Workshop) [Paper] [Code] (Techs - Gated GNN + Meta Learning: MAML, FO-MAML, ANIL)

<i>Active Learning</i>

• ASGN: An Active Semi-supervised Graph Neural Network for Molecular Property Prediction (aka: ASGN; KDD 2020) [Paper] [Code] (Techs - Active Learning)
• Evidential Deep Learning for Guided Molecular Property Prediction and Discovery (NeurIPS 2020 Workshop) [Talk]
• Batched Bayesian Optimization for Drug Design in Noisy Environments (J Chem Inf Model 2022) [Paper] [Code]

5. Addressing Existing Challenges

<i>Model Interpretation</i>

• Drug Discovery Maps, a Machine Learning Model That Visualizes and Predicts Kinome−Inhibitor Interaction Landscapes (J Chem Inf Model 2018) [Paper]
• Using attribution to decode binding mechanism in neural network models for chemistry (PNAS 2019) [Paper]
• Interpretation of QSAR Models by Coloring Atoms According to Changes in Predicted Activity: How Robust Is It? (J Chem Inf Model 2019) [Paper]
• Building of Robust and Interpretable QSAR Classification Models by Means of the Rivality Index (J Chem Inf Model 2019) [Paper]

<i>Dataset Concerns</i>

• In Need of Bias Control: Evaluating Chemical Data for Machine Learning in Structure-Based Virtual Screening (J Chem Inf Model 2019) [Paper]
• Deep Learning-Based Imbalanced Data Classification for Drug Discovery (J Chem Inf Model 2020) [Paper] [Code]

<i>Uncertainty Estimation</i>

• General Approach to Estimate Error Bars for Quantitative Structure−Activity Relationship Predictions of Molecular Activity (J Chem Inf Model 2018) [Paper]
• Assessment and Reproducibility of Quantitative Structure−Activity Relationship Models by the Nonexpert (J Chem Inf Model 2018) [Paper]
• Deep Confidence: A Computationally Efficient Framework for Calculating Reliable Prediction Errors for Deep Neural Networks (J Chem Inf Model 2018) [Paper]
• Reliable Prediction Errors for Deep Neural Networks Using Test-Time Dropout (J Chem Inf Model 2019) [Paper]
• Minimal-uncertainty prediction of general drug-likeness based on Bayesian neural networks (Nat Mach Intell 2020) [Paper]
• Assigning Confidence to Molecular Property Prediction (arXiv 2021) [Paper]
• Gi and Pal Scores: Deep Neural Network Generalization Statistics (ICLR 2021 Workshop) [Paper]
• Probabilistic Random Forest improves bioactivity predictions close to the classification threshold by taking into account experimental uncertainty (J Cheminf 2021) [Paper]

<i>Representation Capacity</i>

• Ligand-Based Virtual Screening Using Graph Edit Distance as Molecular Similarity Measure (J Chem Inf Model 2019) [Paper]
• Optimal Transport Graph Neural Networks (arXiv 2020) [Paper]

<i>Out-of-Distribution Generalization</i>

• Dissecting Machine-Learning Prediction of Molecular Activity: Is an Applicability Domain Needed for Quantitative Structure−Activity Relationship Models Based on Deep Neural Networks? (J Chem Inf Model 2018) [Paper]
• Most Ligand-Based Classification Benchmarks Reward Memorization Rather than Generalization (J Chem Inf Model 2018) [Paper]
• Molecular Similarity-Based Domain Applicability Metric Efficiently Identifies Out-of-Domain Compounds (J Chem Inf Model 2018) [Paper]

<i>Threshold Adjustment</i>

• GHOST: Adjusting the Decision Threshold to Handle Imbalanced Data in Machine Learning (J Chem Inf Model 2021) [Paper] [Code]

<i>Model Comparison</i>

• Validating the validation: reanalyzing a large‑scale comparison of deep learning and machine learning models for bioactivity prediction (J Comput Aided Mol Des 2020) [Paper]
• Comparing classification models-a practical tutorial (J Comput Aided Mol Des 2021) [Paper]

<i>Model Adoption</i>

• A Turing Test for Molecular Generators (J Med Chem 2020) [Paper]

<i>Molecular Docking</i>

• Docking and scoring in virtual screening for drug discovery: methods and applications (Nat Rev Drug Discov 2004) [Paper]
• Benchmarking sets for molecular docking (J Med Chem 2006) [Paper]
• Molecular Docking: A powerful approach for structure-based drug discovery (Curr Comput Aided Drug Des 2011) [Paper]
• Software for Molecular Docking: a review (Biophys Rev 2017) [Paper]
• Deep Docking: A Deep Learning Platform for Augmentation of Structure Based Drug Discovery (ACS Cent Sci 2020) [Paper]
• A Deep-Learning Approach toward Rational Molecular Docking Protocol Selection (Molecules 2020) [Paper]
• GNINA 1.0: molecular docking with deep learning (J Cheminf 2021) [Paper]

<i>Molecular Fragmentation & Assembly</i>

• Molecular generation by Fast Assembly of (Deep)SMILES fragments (J Cheminf 2021) [Paper] [Code]