Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules

2019/06/02 by Niklas W. A. Gebauer, Michael Gastegger, Gebauer, Niklas W. A. +3 · 20 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · Materials Science · #Chemical Physics (physics.chem-ph) #Computational Drug Discovery Methods #Computational Physics (physics.comp-ph) #FOS: Computer and information sciences #FOS: Physical sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning in Materials Science #Protein Structure and Dynamics

paper · pdf · doi:10.48550/arxiv.1906.00957

openalex publication_date 2019/06/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Deep learning has proven to yield fast and accurate predictions of quantum-chemical properties to accelerate the discovery of novel molecules and materials. As an exhaustive exploration of the vast chemical space is still infeasible, we require generative models that guide our search towards systems with desired properties. While graph-based models have previously been proposed, they are restricted by a lack of spatial information such that they are unable to recognize spatial isomerism and non-bonded interactions. Here, we introduce a generative neural network for 3d point sets that respects the rotational invariance of the targeted structures. We apply it to the generation of molecules and demonstrate its ability to approximate the distribution of equilibrium structures using spatial metrics as well as established measures from chemoinformatics. As our model is able to capture the complex relationship between 3d geometry and electronic properties, we bias the distribution of the generator towards molecules with a small HOMO-LUMO gap - an important property for the design of organic solar cells.

Citations

• Adam: A Method for Stochastic Optimization
• Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density
• Accurate spin-dependent electron liquid correlation energies for local spin density calculations: a critical analysis
• A consistent and accurate<i>ab initio</i>parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu
• Density-functional thermochemistry. III. The role of exact exchange
• Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields
• Molecular graph convolutions: moving beyond fingerprints
• Quantum-chemical insights from deep tensor neural networks
• Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
• Neural Message Passing for Quantum Chemistry
• Objective-Reinforced Generative Adversarial Networks (ORGAN) for Sequence Generation Models
• SchNet – A deep learning architecture for molecules and materials
• Junction Tree Variational Autoencoder for Molecular Graph Generation
• Learning Multimodal Graph-to-Graph Translation for Molecular Optimization
• Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation
• Conditional Image Generation with PixelCNN Decoders
• PyTorch: An Imperative Style, High-Performance Deep Learning Library

Cited by

• CoPHo: Classifier-guided Conditional Topology Generation with Persistent Homology
• MolSculpt: Sculpting 3D Molecular Geometries from Chemical Syntax
• Equivariant Diffusion for Crystal Structure Prediction
• NEAT: Neighborhood-Guided, Efficient, Autoregressive Set Transformer for 3D Molecular Generation
• Autoregressive Boltzmann Generators
• Logical GANs: Adversarial Learning through Ehrenfeucht Fraisse Games
• Accelerating Materials Design via LLM-Guided Evolutionary Search
• Uncertainty-Aware Multi-Objective Reinforcement Learning-Guided Diffusion Models for 3D De Novo Molecular Design
• FLOWR.root: A flow matching based foundation model for joint multi-purpose structure-aware 3D ligand generation and affinity prediction
• Controllable Graph Generation with Diffusion Models via Inference-Time Tree Search Guidance
• Predictive Feature Caching for Training-free Acceleration of Molecular Geometry Generation
• Multi-domain Distribution Learning for De Novo Drug Design
• MolMark: Safeguarding Molecular Structures through Learnable Atom-Level Watermarking
• Inverse Design of Amorphous Materials with Targeted Properties
• Guiding Diffusion Models with Reinforcement Learning for Stable Molecule Generation
• SOLD: SELFIES-based Objective-driven Latent Diffusion
• NovoMolGen: Rethinking Molecular Language Model Pretraining
• FlowMol3: Flow Matching for 3D De Novo Small-Molecule Generation
• Data-driven particle dynamics: Structure-preserving coarse-graining for emergent behavior in non-equilibrium systems
• Chem3DLLM: 3D Multimodal Large Language Models for Chemistry

Related

• 3D Molecular Geometry Analysis with 2D Graphs
• Using GNN property predictors as molecule generators
• Using GNN property predictors as molecule generators
• Molecule3D: A Benchmark for Predicting 3D Geometries from Molecular Graphs
• Augmenting Molecular Graphs with Geometries via Machine Learning Interatomic Potentials