Optimization of Molecules via Deep Reinforcement Learning

Abstract

We present a framework, which we call Molecule Deep $Q$-Networks (MolDQN), for molecule optimization by combining domain knowledge of chemistry and state-of-the-art reinforcement learning techniques (double $Q$-learning and randomized value functions). We directly define modifications on molecules, thereby ensuring 100\% chemical validity. Further, we operate without pre-training on any dataset to avoid possible bias from the choice of that set. Inspired by problems faced during medicinal chemistry lead optimization, we extend our model with multi-objective reinforcement learning, which maximizes drug-likeness while maintaining similarity to the original molecule. We further show the path through chemical space to achieve optimization for a molecule to understand how the model works.

References in corpus (2)

• Automatic chemical design using a data-driven continuous representation of molecules
• Deep Reinforcement Learning for De-Novo Drug Design

Cited by in corpus (30)

• Perspective on integrating machine learning into computational chemistry and materials science
• Computer-Aided Multi-Objective Optimization in Small Molecule Discovery
• Using Attribution to Decode Dataset Bias in Neural Network Models for Chemistry
• Quantum Optimization with a Novel Gibbs Objective Function and Ansatz Architecture Search
• A Deep Generative Model for Molecule Optimization via One Fragment Modification
• Predicting Aqueous Solubility of Organic Molecules Using Deep Learning Models with Varied Molecular Representations
• Optimizing Molecules using Efficient Queries from Property Evaluations
• Scaffold-constrained molecular generation
• Molecular Design in Synthetically Accessible Chemical Space via Deep Reinforcement Learning
• Bidirectional Generation of Structure and Properties Through a Single Molecular Foundation Model
• On-the-fly Autonomous Control of Neutron Diffraction via Physics-Informed Bayesian Active Learning
• Deep Molecular Dreaming: Inverse machine learning for de-novo molecular design and interpretability with surjective representations
• Ten Quick Tips for Deep Learning in Biology
• Graph Machine Learning for Design of High-Octane Fuels
• Multi-Objective Latent Space Optimization of Generative Molecular Design Models
• Augmentation of Universal Potentials for Broad Applications
• Cascading symmetry constraint during machine learning-enabled structural search for sulfur induced Cu(111)-$(\sqrt{43}\times \sqrt{43})$ surface reconstruction
• Reversible Upper Confidence Bound Algorithm to Generate Diverse Optimized Candidates
• GraphXForm: Graph transformer for computer-aided molecular design
• Molecular Dipole Moment Learning via Rotationally Equivariant Gaussian Process Regression with Derivatives in Molecular-orbital-based Machine Learning
• Optimizing thermodynamic trajectories using evolutionary and gradient-based reinforcement learning
• Evolutionary Monte Carlo of QM properties in chemical space: Electrolyte design
• G-PCGRL: Procedural Graph Data Generation via Reinforcement Learning
• Human-level molecular optimization driven by mol-gene evolution
• Leveraging Active Subspaces to Capture Epistemic Model Uncertainty in Deep Generative Models for Molecular Design
• Atomistic Global Optimization X: A Python package for optimization of atomistic structures
• Multi-Constraint Molecular Generation using Sparsely Labelled Training Data for Localized High-Concentration Electrolyte Diluent Screening
• Quantum Chemistry Driven Molecular Inverse Design with Data-free Reinforcement Learning
• Curiosity as a Self-Supervised Method to Improve Exploration in De novo Drug Design
• Design of Detectors at the Electron Ion Collider with Artificial Intelligence