mlsanityData to 14 Oct 2026

Structure-based drug design by denoising voxel grids

arXiv 2405.03961 ↗First posted 7 May 2024cs.LG q-bio.BMcited by 19

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

We present VoxBind, a new score-based generative model for 3D molecules conditioned on protein structures. Our approach represents molecules as 3D atomic density grids and leverages a 3D voxel-denoising network for learning and generation. We extend the neural empirical Bayes formalism (Saremi & Hyvarinen, 2019) to the conditional setting and generate structure-conditioned molecules with a two-step procedure: (i) sample noisy molecules from the Gaussian-smoothed conditional distribution with underdamped Langevin MCMC using the learned score function and (ii) estimate clean molecules from the noisy samples with single-step denoising. Compared to the current state of the art, our model is simpler to train, significantly faster to sample from, and achieves better results on extensive in silico benchmarks -- the generated molecules are more diverse, exhibit fewer steric clashes, and bind with higher affinity to protein pockets. The code is available at https://github.com/genentech/voxbind/.

Key facts. Cited by 19 papers in the map (data through 14 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 41 of its 75 references that are in the map. First posted on arXiv 7 May 2024.

Builds on

41 of 75 references.

Closest ancestors · 6

Same direction, newest first

  1. 2024DecompDiff: Diffusion Models with Decomposed Priors for Structure-Based Drug Design26 Feb
  2. 2023Geometric Latent Diffusion Models for 3D Molecule Generation2 May
  3. 2023MiDi: Mixed Graph and 3D Denoising Diffusion for Molecule Generation17 Feb
  4. 2022Structure-based Drug Design with Equivariant Diffusion Models24 Oct
  5. 2022MDM: Molecular Diffusion Model for 3D Molecule Generation13 Sep
  6. 2022Equivariant Diffusion for Molecule Generation in 3D31 Mar

Other references · 35

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  2. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  3. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  4. 2018How Powerful are Graph Neural Networks?cited by 4,393
  5. 2013Auto-Encoding Variational Bayescited by 4,053
  6. 2015Variational Inference with Normalizing Flowscited by 2,170
  7. 2021Image Super-Resolution via Iterative Refinementcited by 1,511
  8. 2021Palette: Image-to-Image Diffusion Modelscited by 1,300
  9. 2021E(n) Equivariant Graph Neural Networkscited by 965
  10. 2017Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learningcited by 950
  11. 2018Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networkscited by 881
  12. 2022MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fieldscited by 474
23 more
  1. 20183D Steerable CNNs: Learning Rotationally Equivariant Features in Volumetric Datacited by 354
  2. 2021SegDiff: Image Segmentation with Diffusion Probabilistic Modelscited by 280
  3. 2020Equivariant Flows: Exact Likelihood Generative Learning for Symmetric Densitiescited by 250
  4. 2021Are Transformers More Robust Than CNNs?cited by 187
  5. 2022Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pocketscited by 183
  6. 20233D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Predictioncited by 177
  7. 2022e3nn: Euclidean Neural Networkscited by 174
  8. 2019Symmetry-adapted generation of 3d point sets for the targeted discovery of moleculescited by 172
  9. 2022A 3D Generative Model for Structure-Based Drug Designcited by 149
  10. 2023The Surprising Effectiveness of Diffusion Models for Optical Flow and Monocular Depth Estimationcited by 116
  11. 2022Generating 3D Molecules for Target Protein Bindingcited by 104
  12. 2020ATOM3D: Tasks On Molecules in Three Dimensionscited by 74
  13. 2021Generating 3D Molecules Conditional on Receptor Binding Sites with Deep Generative Modelscited by 73
  14. 2023Language models can generate molecules, materials, and protein binding sites directly in three dimensions as XYZ, CIF, and PDB filescited by 59
  15. 2022The Lie Derivative for Measuring Learned Equivariancecited by 54
  16. 2019Neural Empirical Bayescited by 46
  17. 20233D molecule generation by denoising voxel gridscited by 40
  18. 2023Protein Discovery with Discrete Walk-Jump Samplingcited by 38
  19. 2020Learning a Continuous Representation of 3D Molecular Structures with Deep Generative Modelscited by 24
  20. 2022Zero-Shot 3D Drug Design by Sketching and Generatingcited by 24
  21. 2022Equivariant Shape-Conditioned Generation of 3D Molecules for Ligand-Based Drug Designcited by 19
  22. 2018Generating equilibrium molecules with deep neural networkscited by 10
  23. 2023Universal Smoothed Score Functions for Generative Modelingcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 9; 2026: 7.

19 papers.

Sort:

By area:

  1. 2024CBGBench: Fill in the Blank of Protein-Molecule Complex Binding Graphcited by 19
  2. 2025UniMoMo: Unified Generative Modeling of 3D Molecules for De Novo Binder Designcited by 12
  3. 2024Empower Structure-Based Molecule Optimization with Gradient Guided Bayesian Flow Networkscited by 6
  4. 2024JAMUN: Bridging Smoothed Molecular Dynamics and Score-Based Learning for Conformational Ensemblescited by 4
  5. 2025Unified all-atom molecule generation with neural fieldscited by 3
  6. 2025Do we need equivariant models for molecule generation?cited by 2
  7. 2026Generative Molecular Morphing for Flexible-Size Design via Unbalanced Optimal Transportcited by 2
  8. 2025IBEX: Information-Bottleneck-EXplored Coarse-to-Fine Molecular Generation under Limited Datacited by 1
  9. 2025Beyond Atoms: Evaluating Electron Density Representation for 3D Molecular Learningcited by 1
  10. 2025From Macro to Micro: Benchmarking Microscopic Spatial Intelligence on Molecules via Vision-Language Modelscited by 1
  11. 2026MolSafeEval: A Benchmark for Uncovering Safety Risks in AI-Generated Moleculescited by 1
  12. 2025TransDiffSBDD: Causality-Aware Multi-Modal Structure-Based Drug Designcited by 0
7 more
  1. 2025MSCoD: An Enhanced Bayesian Updating Framework with Multi-Scale Information Bottleneck and Cooperative Attention for Structure-Based Drug Designcited by 0
  2. 2025SculptDrug : A Spatial Condition-Aware Bayesian Flow Model for Structure-based Drug Designcited by 0
  3. 2026Edge-aware GAT-based protein binding site predictioncited by 0
  4. 2026Multigrid Training for Molecular Generation using Graph Neural Networkscited by 0
  5. 2026Sesame: Structure-Aware Molecular Generation via Spatial Density-Map Conditioningcited by 0
  6. 2026A Systematic Evaluation of Molecule Generation Models for De Novo Drug Design: From Benchmarks to Practical Insightscited by 0
  7. 2026Blending experimental electron density information to protein-ligand interactioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022DiffBP: Generative Diffusion of 3D Molecules for Target Protein Binding
  2. 2024NEBULA: Neural Empirical Bayes Under Latent Representations for Efficient and Controllable Design of Molecular Libraries
  3. 2023Geometry-Complete Diffusion for 3D Molecule Generation and Optimization
  4. 2023Generating Novel, Designable, and Diverse Protein Structures by Equivariantly Diffusing Oriented Residue Clouds
  5. 2024Diffusion Models in textitDe Novo Drug Design
  6. 2022Protein Structure and Sequence Generation with Equivariant Denoising Diffusion Probabilistic Models
  7. 2024Mask prior-guided denoising diffusion improves inverse protein folding
  8. 2024ControlMol: Adding Substructure Control To Molecule Diffusion Models