Multi-level Self-supervised Pretraining on Compositional Hierarchical Graph for Molecular Property Prediction

• URL: http://arxiv.org/abs/2605.16088v1
• Date: Fri, 15 May 2026 15:47:11 GMT
• Title: Multi-level Self-supervised Pretraining on Compositional Hierarchical Graph for Molecular Property Prediction
• Abstract summary: MolCHG is a multi-level self-supervised pretraining framework built upon a novel Compositional Hierarchical Graph.<n>Our architecture elevates bond-level information to independently evolving node representations.<n>Experiments on nine MoleculeNet benchmarks demonstrate that MolCHG achieves the best performance on seven datasets.
• Score: 0.5097809301149341
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Self-supervised pretraining on molecular graphs has emerged as a promising approach for molecular property prediction, yet most existing methods operate at a single structural granularity and treat bond information as auxiliary edge attributes rather than as an independent semantic layer. In this work, we propose MolCHG, a multi-level self-supervised pretraining framework built upon a novel Compositional Hierarchical Graph that organizes molecular structure into four types of nodes across three semantic levels. By introducing a bond graph that operates in parallel with the atom graph, our architecture elevates bond-level information to independently evolving node representations, enabling fragment nodes to aggregate atom-level and bond-level semantics on an equal footing. We design three level-specific pretraining objectives: an atom-bond cross-view contrastive task that aligns the atom-view and bond-view representations within each fragment, a fragment-level functional group prediction task to inject domain-relevant chemical knowledge, and graph-level structure prediction tasks to encode global molecular topology. Experiments on nine MoleculeNet benchmarks demonstrate that MolCHG achieves the best performance on seven datasets across both classification and regression tasks, remaining competitive with the strongest baselines on the rest. Ablation studies further confirm that the multi-level supervision signals are complementary and that each component contributes to the overall performance.

Related papers

• Multi-View Molecular Representation Learning with Hierarchical Graphs and Contextualized Fingerprints [2.7742696928124917]
  HiFi-Mol is a multi-view framework that pretrains a hierarchical graph encoder and a contextualized fingerprint encoder before downstream integration.<n>During fine-tuning, HiFi-Mol combines projected multi-resolution graph features with fingerprint embeddings for downstream prediction.
  arXiv  Detail & Related papers  (2026-09-14T14:07:19Z)
• Beyond the Academic Monoculture: A Unified Framework and Industrial Perspective for Attributed Graph Clustering [75.50670592447219]
  Attributed Graph Clustering (AGC) is a fundamental unsupervised task that partitions nodes into cohesive groups by jointly modeling structural topology and node attributes.<n>This survey provides a comprehensive, industrially grounded review of AGC from three complementary perspectives.
  arXiv  Detail & Related papers  (2026-03-21T14:15:34Z)
• Hierarchical Molecular Representation Learning via Fragment-Based Self-Supervised Embedding Prediction [7.459632891054827]
  Graph self-supervised learning (GSSL) has demonstrated strong potential for generating expressive graph embeddings without the need for human annotations.<n>GraSPNet is a hierarchical self-supervised framework that explicitly models both atomic-level and fragment-level semantics.<n>GraSPNet learns chemically meaningful representations and consistently outperforms state-of-the-art GSSL methods in transfer learning settings.
  arXiv  Detail & Related papers  (2026-02-23T20:41:44Z)
• ProtoMol: Enhancing Molecular Property Prediction via Prototype-Guided Multimodal Learning [14.289447310645878]
  ProtoMol is a prototype-guided framework that enables fine-grained integration and consistent semantic alignment between modalities.<n>ProtoMol consistently outperforms state-of-the-art baselines across a variety of molecular property prediction tasks.
  arXiv  Detail & Related papers  (2025-10-19T13:19:37Z)
• GeoRecon: Graph-Level Representation Learning for 3D Molecules via Reconstruction-Based Pretraining [34.23512730854597]
  We introduce GeoRecon, a graph-level pretraining framework that shifts the focus from individual atoms to the molecule as an integrated whole.<n>GeoRecon formulates a graph-level reconstruction task that guides geometry reconstruction while inducing smoother and more transferable latent spaces.<n>Without relying on external supervision, GeoRecon generally improves over backbone baselines on multiple molecular benchmarks.
  arXiv  Detail & Related papers  (2025-06-16T07:35:49Z)
• Hi-GMAE: Hierarchical Graph Masked Autoencoders [90.30572554544385]
  Hierarchical Graph Masked AutoEncoders (Hi-GMAE) Hi-GMAE is a novel multi-scale GMAE framework designed to handle the hierarchical structures within graphs. Our experiments on 15 graph datasets consistently demonstrate that Hi-GMAE outperforms 17 state-of-the-art self-supervised competitors.
  arXiv  Detail & Related papers  (2024-05-17T09:08:37Z)
• Atomas: Hierarchical Alignment on Molecule-Text for Unified Molecule Understanding and Generation [42.08917809689811]
  Cross-modal representation learning has emerged as a promising direction for enhancing the quality of molecular representation.<n>We propose Atomas, a hierarchical molecular representation learning framework that jointly learns representations from SMILES strings and text.<n>Atomas achieves superior performance across 12 tasks on 11 datasets, outperforming 11 baseline models.
  arXiv  Detail & Related papers  (2024-04-23T12:35:44Z)
• Incorporating Higher-order Structural Information for Graph Clustering [6.027366081402081]
  Graph convolutional network (GCN) has emerged as a powerful tool for deep clustering. We propose a novel graph clustering network to make full use of graph structural information. Our proposed model outperforms many state-of-the-art methods on various datasets.
  arXiv  Detail & Related papers  (2024-03-17T04:42:41Z)
• Bi-level Contrastive Learning for Knowledge-Enhanced Molecule Representations [68.32093648671496]
  We introduce GODE, which accounts for the dual-level structure inherent in molecules.<n> Molecules possess an intrinsic graph structure and simultaneously function as nodes within a broader molecular knowledge graph.<n>By pre-training two GNNs on different graph structures, GODE effectively fuses molecular structures with their corresponding knowledge graph substructures.
  arXiv  Detail & Related papers  (2023-06-02T15:49:45Z)
• Self-supervised Graph-level Representation Learning with Local and Global Structure [71.45196938842608]
  We propose a unified framework called Local-instance and Global-semantic Learning (GraphLoG) for self-supervised whole-graph representation learning. Besides preserving the local similarities, GraphLoG introduces the hierarchical prototypes to capture the global semantic clusters. An efficient online expectation-maximization (EM) algorithm is further developed for learning the model.
  arXiv  Detail & Related papers  (2021-06-08T05:25:38Z)
• Distance-aware Molecule Graph Attention Network for Drug-Target Binding Affinity Prediction [54.93890176891602]
  We propose a diStance-aware Molecule graph Attention Network (S-MAN) tailored to drug-target binding affinity prediction. As a dedicated solution, we first propose a position encoding mechanism to integrate the topological structure and spatial position information into the constructed pocket-ligand graph. We also propose a novel edge-node hierarchical attentive aggregation structure which has edge-level aggregation and node-level aggregation.
  arXiv  Detail & Related papers  (2020-12-17T17:44:01Z)
• Spectral Embedding of Graph Networks [76.27138343125985]
  We introduce an unsupervised graph embedding that trades off local node similarity and connectivity, and global structure. The embedding is based on a generalized graph Laplacian, whose eigenvectors compactly capture both network structure and neighborhood proximity in a single representation.
  arXiv  Detail & Related papers  (2020-09-30T04:59:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.