mlsanityData to 7 Oct 2026

h-MINT: Modeling Pocket-Ligand Binding with Hierarchical Molecular Interaction Network

arXiv 2604.23134 ↗First posted 25 Apr 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Accurate molecular representations are critical for drug discovery, and a central challenge lies in capturing the chemical environment of molecular fragments, as key interactions, such as H-bond and π stacking, occur only under specific local conditions. Most existing approaches represent molecules as atom-level graphs; however, atom-level representations can hardly express higher-order chemical context (e.g., stereochemistry, lone pairs, conjugation). Fragment-based methods (e.g., principal subgraph, predefined functional groups) fail to preserve essential information such as chirality, aromaticity, and ionic states. This work addresses these limitations from two aspects. (i) OverlapBPE tokenization. We propose a novel data-driven molecule tokenization method. Unlike existing approaches, our method allows overlapping fragments, reflecting the inherently fuzzy boundaries of small-molecule substructures and, together with enriched chemical information at the token level, thereby preserving a more complete chemical context. (ii) h-MINT model. OverlapBPE induces many-to-many atom-fragment mappings, which necessitate a new hierarchical architecture. We therefore develop a hierarchical molecular interaction network capable of jointly modeling interactions at both atom and fragment levels. By supporting fragment overlaps, the model naturally accommodates the many-to-many atom-fragment mappings introduced by the OverlapBPE scheme. Extensive evaluation against state-of-the-art methods shows our method improves binding affinity prediction by 2-4% Pearson/Spearman correlation on PDBBind and LBA, enhances virtual screening by 1-3% in key metrics on DUD-E and LIT-PCBA, and achieves the best overall HTS performance on PubChem assays. Further analysis demonstrates that our method effectively captures interactive information while maintaining good generalization.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 40 of its 75 references that are in the map. First posted on arXiv 25 Apr 2026.

Builds on

40 of 75 references.

References · 40

  1. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  2. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,898
  3. 2021E(n) Equivariant Graph Neural Networkscited by 951
  4. 2022Equivariant Diffusion for Molecule Generation in 3Dcited by 666
  5. 2018Junction Tree Variational Autoencoder for Molecular Graph Generationcited by 647
  6. 2017SchNet: A continuous-filter convolutional neural network for modeling quantum interactionscited by 569
  7. 2022GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generationcited by 511
  8. 2022MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fieldscited by 467
  9. 2021GemNet: Universal Directional Graph Neural Networks for Moleculescited by 342
  10. 2020Learning from Protein Structure with Geometric Vector Perceptronscited by 308
  11. 2022Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphscited by 258
  12. 2020Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Moleculescited by 254
28 more
  1. 2019Evaluating Protein Transfer Learning with TAPEcited by 229
  2. 2020Hierarchical Generation of Molecular Graphs using Structural Motifscited by 204
  3. 2022TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentialscited by 179
  4. 2018DeepDTA: Deep Drug-Target Binding Affinity Predictioncited by 171
  5. 2021Spherical Message Passing for 3D Graph Networkscited by 166
  6. 20213D Infomax improves GNNs for Molecular Property Predictioncited by 146
  7. 2022Pre-training via Denoising for Molecular Property Predictioncited by 124
  8. 2021Geometric Deep Learning on Molecular Representationscited by 113
  9. 2022Protein Structure and Sequence Generation with Equivariant Denoising Diffusion Probabilistic Modelscited by 97
  10. 2022PiFold: Toward effective and efficient protein inverse foldingcited by 82
  11. 2022Conditional Antibody Design as 3D Equivariant Graph Translationcited by 75
  12. 2020ATOM3D: Tasks On Molecules in Three Dimensionscited by 73
  13. 2021Hit and Lead Discovery with Explorative RL and Fragment-based Molecule Generationcited by 68
  14. 2022Learning Hierarchical Protein Representations via Complete 3D Graph Networkscited by 54
  15. 2021Equivariant Graph Neural Networks for 3D Macromolecular Structurecited by 53
  16. 2021Molecule Generation by Principal Subgraph Mining and Assemblingcited by 53
  17. 2022Multi-Scale Representation Learning on Proteinscited by 52
  18. 2019Learning protein sequence embeddings using information from structurecited by 50
  19. 2022KerGNNs: Interpretable Graph Neural Networks with Graph Kernelscited by 50
  20. 2023A new perspective on building efficient and expressive 3D equivariant graph neural networkscited by 46
  21. 2023De Novo Molecular Generation via Connection-aware Motif Miningcited by 45
  22. 2021Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotationscited by 29
  23. 2023DrugCLIP: Contrastive Protein-Molecule Representation Learning for Virtual Screeningcited by 28
  24. 2023Generalist Equivariant Transformer Towards 3D Molecular Interaction Learningcited by 18
  25. 2022Antibody-Antigen Docking and Design via Hierarchical Equivariant Refinementcited by 15
  26. 2023ProFSA: Self-supervised Pocket Pretraining via Protein Fragment-Surroundings Alignmentcited by 12
  27. 2023Protein-ligand binding representation learning from fine-grained interactionscited by 12
  28. 2024GraphBPE: Molecular Graphs Meet Byte-Pair Encodingcited by 7

Built on by

Not cited yet.