awesome-small-molecule-ml
A curated list of resources for machine learning for small-molecule drug discovery
https://github.com/benb111/awesome-small-molecule-ml
Last synced: 1 day ago
JSON representation
Blogs
Data sets
Frameworks, Libraries, and Software Tools
Papers
ADME and toxicity prediction
- A Graph Neural Network Approach to Molecule Carcinogenicity Prediction - .
- CardioTox net: a robust predictor for hERG channel blockade based on deep learning meta-feature ensembles - . [Code]
- Validating ADME QSAR Models Using Marketed Drugs - .
- Deep Learning-Based Prediction of Drug-Induced Cardiotoxicity - . [Code]
- In Silico Absorption, Distribution, Metabolism, Excretion, and Pharmacokinetics (ADME-PK): Utility and Best Practices - .
- DeepHIT: a deep learning framework for prediction of hERG-induced cardiotoxicity - . [Code]
- Support Vector Machine model for hERG inhibitory activities based on the integrated hERG database using descriptor selection by NSGA-II - . [Data]
DNA-encoded libraries (DELs)
- Machine Learning on DNA-Encoded Library Count Data Using an Uncertainty-Aware Probabilistic Loss Function - . [Code]
- Machine Learning on DNA-Encoded Libraries: A New Paradigm for Hit Finding - .
Generative algorithms
- Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation - . [Code]
- Molecular generation by Fast Assembly of (Deep)SMILES fragments - . [Code]
- Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Design - . [Code]
- Deep Generative Models for 3D Linker Design - . [Code]
- Hierarchical Generation of Molecular Graphs using Structural Motifs - . [Code]
- CReM: chemically reasonable mutations framework for structure generation - . [Code]
- GuacaMol: Benchmarking Models for de Novo Molecular Design - . [Code]
- Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation - . [Code]
- Optimization of Molecules via Deep Reinforcement Learning - . [Code (official version)] [PyTorch implementation]
- Junction Tree Variational Autoencoder for Molecular Graph Generation - . [Code]
- De Novo Design of Bioactive Small Molecules by Artificial Intelligence - .
- R-group replacement database for medicinal chemistry - .
- MolecularRNN: Generating realistic molecular graphs with optimized properties - .
Hit finding and potency prediciton
- EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction - . [Code]
- DOCKSTRING: easy molecular docking yields better benchmarks for ligand design - . [Code] [Data]
- Accelerating high-throughput virtual screening through molecular pool-based active learning - . [Code]
- Deep Docking: A Deep Learning Platform for Augmentation of Structure Based Drug Discovery - . [Code]
- Adding Stochastic Negative Examples into Machine Learning Improves Molecular Bioactivity Prediction - .
- Ultra-large library docking for discovering new chemotypes - .
- A practical guide to large-scale docking - .
MS/MS prediction
- MassFormer: Tandem Mass Spectrum Prediction for Small Molecules using Graph Transformers - . [Code]
- Prefix-Tree Decoding for Predicting Mass Spectra from Molecules - . [Code]
- 3DMolMS: prediction of tandem mass spectra from 3D molecular conformations - . [Code]
- CFM-ID 4.0: More Accurate ESI-MS/MS Spectral Prediction and Compound Identification - .
- Rapid Prediction of Electron–Ionization Mass Spectrometry Using Neural Networks - . [Code]
Representation, transfer learning, and few-shot learning
- SELFIES and the future of molecular string representations - .
- ChemBERTa-2: Towards Chemical Foundation Models - . [Code]
- E(n) Equivariant Graph Neural Networks - . [Code]
- FS-Mol: A Few-Shot Learning Dataset of Molecules - . [Code]
- ATOM3D: Tasks On Molecules in Three Dimensions - .
- X-MOL: large-scale pre-training for molecular understanding and diverse molecular analysis - . [Code]
- Do Transformers Really Perform Bad for Graph Representation? (Graphormer paper) - . [Code]
- Attention-Based Learning on Molecular Ensembles - .
- Inductive transfer learning for molecular activity prediction: Next-Gen QSAR Models with MolPMoFiT - . [Code]
- Self-Supervised Graph Transformer on Large-Scale Molecular Data (GROVER paper) - . [Code]
- Strategies for Pre-training Graph Neural Networks - . [Code]
- Analyzing Learned Molecular Representations for Property Prediction (Chemprop) - . [Code]
- PotentialNet for Molecular Property Prediction - .
- Low Data Drug Discovery with One-Shot Learning - .
- Molecule Attention Transformer - . [Code]
- Molecular Contrastive Learning of Representations via Graph Neural Networks - . [Code]
- Meta-Learning GNN Initializations for Low-Resource Molecular Property Prediction - [Code]
Survey papers and books
- Critical assessment of AI in drug discovery - .
- Deep Learning for Molecules and Materials - .
- Defining and Exploring Chemical Spaces - .
- Learning Molecular Representations for Medicinal Chemistry - .
- Applications of Deep Learning in Molecule Generation and Molecular Property Prediction - .
- Transfer Learning for Drug Discovery - .
Synthetic accessability and retrosynthetic planning
- Data augmentation and pretraining for template-based retrosynthetic prediction in computer-aided synthesis planning - .
- Reinforcement Learning for Bioretrosynthesis - .
- Learning Graph Models for Retrosynthesis Prediction - .
- Retrosynthesis Prediction with Conditional Graph Logic Network - . [Code]
- SCScore: Synthetic Complexity Learned from a Reaction Corpus - . [Code] [DeepChem implementation]
Visualization and interpretability
- ChemInformatics Model Explorer (CIME): Exploratory analysis of chemical model explanations - . [Code]
- Benchmarks for interpretation of QSAR models - . [Code]
- Integrating the Structure–Activity Relationship Matrix Method with Molecular Grid Maps and Activity Landscape Models for Medicinal Chemistry Applications - .
- Finding Constellations in Chemical Space Through Core Analysis - .