mlsanityData to 7 Oct 2026

Directional Message Passing on Molecular Graphs via Synthetic Coordinates

arXiv 2111.04718 ↗First posted 8 Nov 2021cs.LG physics.chem-ph physics.comp-phcited by 24

Not on the map. Why?

Abstract (arXiv)

Graph neural networks that leverage coordinates via directional message passing have recently set the state of the art on multiple molecular property prediction tasks. However, they rely on atom position information that is often unavailable, and obtaining it is usually prohibitively expensive or even impossible. In this paper we propose synthetic coordinates that enable the use of advanced GNNs without requiring the true molecular configuration. We propose two distances as synthetic coordinates: Distance bounds that specify the rough range of molecular configurations, and graph-based distances using a symmetric variant of personalized PageRank. To leverage both distance and angular information we propose a method of transforming normal graph neural networks into directional MPNNs. We show that with this transformation we can reduce the error of a normal graph neural network by 55% on the ZINC benchmark. We furthermore set the state of the art on ZINC and coordinate-free QM9 by incorporating synthetic coordinates in the SMP and DimeNet++ models. Our implementation is available online.

Key facts. Cited by 24 papers in the map (data through 7 Oct 2026). Builds on 37 of its 81 references that are in the map. First posted on arXiv 8 Nov 2021.

Builds on

37 of 81 references.

References · 37

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Graph Attention Networkscited by 8,854
  3. 2018How Powerful are Graph Neural Networks?cited by 4,355
  4. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  5. 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,414
  6. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,148
  7. 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,085
  8. 2013Spectral Networks and Locally Connected Networks on Graphscited by 1,097
  9. 2017MoleculeNet: A Benchmark for Molecular Machine Learningcited by 1,037
  10. 2018Predict then Propagate: Graph Neural Networks meet Personalized PageRankcited by 987
  11. 2021E(n) Equivariant Graph Neural Networkscited by 951
  12. 2018Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networkscited by 878
25 more
  1. 2015Convolutional Networks on Graphs for Learning Molecular Fingerprintscited by 751
  2. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 735
  3. 2020Benchmarking Graph Neural Networkscited by 672
  4. 2018Junction Tree Variational Autoencoder for Molecular Graph Generationcited by 647
  5. 2020On the Bottleneck of Graph Neural Networks and its Practical Implicationscited by 624
  6. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 602
  7. 2017SchNet: A continuous-filter convolutional neural network for modeling quantum interactionscited by 569
  8. 2020Directional Message Passing for Molecular Graphscited by 556
  9. 2020Principal Neighbourhood Aggregation for Graph Netscited by 477
  10. 2021Equivariant message passing for the prediction of tensorial properties and molecular spectracited by 438
  11. 2019Analyzing Learned Molecular Representations for Property Predictioncited by 373
  12. 2017Grammar Variational Autoencodercited by 283
  13. 2019Cormorant: Covariant Molecular Neural Networkscited by 262
  14. 2020Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Moleculescited by 254
  15. 2020DeeperGCN: All You Need to Train Deeper GCNscited by 249
  16. 2020Generalization and Representational Limits of Graph Neural Networkscited by 217
  17. 2020Hierarchical Generation of Molecular Graphs using Structural Motifscited by 204
  18. 2020Directional Graph Networkscited by 132
  19. 2016Discriminative Embeddings of Latent Variable Models for Structured Datacited by 130
  20. 2020Building powerful and equivariant graph neural networks with structural message-passingcited by 95
  21. 2020On the Universality of Rotation Equivariant Point Cloud Networkscited by 80
  22. 2019GNN-FiLM: Graph Neural Networks with Feature-wise Linear Modulationcited by 75
  23. 2020OrbNet: Deep Learning for Quantum Chemistry Using Symmetry-Adapted Atomic-Orbital Featurescited by 63
  24. 2020Hierarchical Inter-Message Passing for Learning on Molecular Graphscited by 59
  25. 2020ASGN: An Active Semi-supervised Graph Neural Network for Molecular Property Predictioncited by 58

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 5; 2023: 7; 2024: 7; 2025: 4; 2026: 1.

24 papers.

Sort:

By area:

  1. 2022Multimodal learning with graphscited by 54
  2. 2022Graph-based Molecular Representation Learningcited by 51
  3. 2023Towards Multi-Layered 3D Garments Animationcited by 16
  4. 2023Spatial Attention Kinetic Networks with E(n)-Equivariancecited by 11
  5. 2025Long-range Brain Graph Transformercited by 11
  6. 2024E(3)-Equivariant Mesh Neural Networkscited by 9
  7. 2025Accurate Ab-initio Neural-network Solutions to Large-Scale Electronic Structure Problemscited by 8
  8. 2022Invariance-Aware Randomized Smoothing Certificatescited by 7
  9. 2024Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Spacecited by 7
  10. 2023Knowledge-augmented Graph Machine Learning for Drug Discovery: A Surveycited by 6
  11. 2024Learning Equivariant Non-Local Electron Density Functionalscited by 6
  12. 2022REMuS-GNN: A Rotation-Equivariant Model for Simulating Continuum Dynamicscited by 3
12 more
  1. 2022Domain-informed graph neural networks: a quantum chemistry case studycited by 3
  2. 2023DuEqNet: Dual-Equivariance Network in Outdoor 3D Object Detection for Autonomous Drivingcited by 3
  3. 2025AI for Scientific Discovery is a Social Problemcited by 3
  4. 20233D Molecular Geometry Analysis with 2D Graphscited by 1
  5. 2024Enhancing Molecular Property Prediction with Auxiliary Learning and Task-Specific Adaptationcited by 1
  6. 2024Dumpling GNN: Hybrid GNN Enables Better ADC Payload Activity Prediction Based on Chemical Structurecited by 1
  7. 2024Grothendieck Graph Neural Networks Framework: An Algebraic Platform for Crafting Topology-Aware GNNscited by 1
  8. 2023On the Interplay of Subset Selection and Informed Graph Neural Networkscited by 0
  9. 2023Differentially Private Graph Neural Network with Importance-Grained Noise Adaptioncited by 0
  10. 2024HEroBM: a deep equivariant graph neural network for universal backmapping from coarse-grained to all-atom representationscited by 0
  11. 2025Hybrid Quantum--Classical Machine Learning Potential with Variational Quantum Circuitscited by 0
  12. 2026Scalar-pathway fidelity improves physical accuracy in short-range equivariant interatomic potentialscited by 0