Generating equilibrium molecules with deep neural networks

Abstract

Discovery of atomistic systems with desirable properties is a major challenge in chemistry and material science. Here we introduce a novel, autoregressive, convolutional deep neural network architecture that generates molecular equilibrium structures by sequentially placing atoms in three-dimensional space. The model estimates the joint probability over molecular configurations with tractable conditional probabilities which only depend on distances between atoms and their nuclear charges. It combines concepts from state-of-the-art atomistic neural networks with auto-regressive generative models for images and speech. We demonstrate that the architecture is capable of generating molecules close to equilibrium for constitutional isomers of C${}_7$O${}_2$H${}_{10}$.

References in corpus (13)

• Neural Message Passing for Quantum Chemistry
• WaveNet: A Generative Model for Raw Audio
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• SchNet - a deep learning architecture for molecules and materials
• Pixel Recurrent Neural Networks
• By-passing the Kohn-Sham equations with machine learning
• Machine learning prediction errors better than DFT accuracy
• How to represent crystal structures for machine learning: towards fast prediction of electronic properties
• Syntax-Directed Variational Autoencoder for Structured Data
• Grammar Variational Autoencoder
• In silico generation of novel, drug-like chemical matter using the LSTM neural network

Cited by in corpus (13)

• Machine Learning Force Fields
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• Inverse design of 3d molecular structures with conditional generative neural networks
• Generative Models for Automatic Chemical Design
• Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules
• Autonomous Reaction Network Exploration in Homogeneous and Heterogeneous Catalysis
• Machine Learning Prediction of Accurate Atomization Energies of Organic Molecules from Low-Fidelity Quantum Chemical Calculations
• Generating valid Euclidean distance matrices
• Reinforcement Learning for Molecular Design Guided by Quantum Mechanics
• Materials property prediction using symmetry-labeled graphs as atomic-position independent descriptors
• Learning a Continuous Representation of 3D Molecular Structures with Deep Generative Models
• Symmetry-Aware Actor-Critic for 3D Molecular Design
• MOLUCINATE: A Generative Model for Molecules in 3D Space