What drives performance in molecular MPNNs? An operator-level factorial benchmark

• URL: http://arxiv.org/abs/2605.30195v1
• Date: Thu, 28 May 2026 16:34:53 GMT
• Title: What drives performance in molecular MPNNs? An operator-level factorial benchmark
• Abstract summary: Message-passing neural networks (MPNNs) are widely used for molecular property prediction, but their deployment as monolithic architectures makes it difficult to identify how specific message-passing operators affect performance.<n>We present an operator-level factorial benchmark that decomposes 2D molecular MPNNs into the three families of message-seed initialization, node-edge fusion, and node update operators.<n>Across this controlled design, performance variation is associated primarily with message construction rather than update complexity.
• Score: 5.551252676792114
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Message-passing neural networks (MPNNs) are widely used for molecular property prediction, but their deployment as monolithic architectures makes it difficult to identify how specific message-passing operators affect performance. We present an operator-level factorial benchmark that decomposes 2D molecular MPNNs into the three families of message-seed initialization, node-edge fusion, and node update operators. The resulting 84 configurations are benchmarked on ten MoleculeNet datasets under a shared experimental setup and statistical analysis protocol. Across this controlled design, performance variation is associated primarily with message construction rather than update complexity. Message-seed initialization shows significant family-level effects for both regression and classification, node-edge fusion shows a significant family-level effect for regression with descriptive advantages for concatenation-based mixing, and the update family shows no statistically supported effect for either endpoint family. A representation probe into the Quinethazone molecule further demonstrates that concatenation-based mixing can better differentiate chemically distinct heteroatoms and withstand oversmoothing than Hadamard gating. Representative configurations selected separately for classification and regression recover competitive performance relative to established molecular graph neural network (GNN) baselines, ranking numerically best on eight of ten benchmark datasets. These empirical results are interpreted through concise mechanistic analyses of representative node-edge fusion and update operators. Our findings provide empirical design heuristics for molecular MPNNs by turning model design from a search over monolithic architectures into a targeted assessment of where and how chemical information enters the message-passing pipeline.

Related papers

• Geometric Multi-color Message Passing Graph Neural Networks for Blood-brain Barrier Permeability Prediction [1.488392495573075]
  This paper introduces the geometric multi-color message-passing graph neural network (GMC-MPNN)<n>Our model constructs weighted colored subgraphs based on atom types to capture the spatial relationships and chemical context that govern blood-brain barrier permeability.
  arXiv  Detail & Related papers  (2025-07-25T03:38:46Z)
• Aligned Manifold Property and Topology Point Clouds for Learning Molecular Properties [55.2480439325792]
  This work introduces AMPTCR, a molecular surface representation that combines local quantum-derived scalar fields and custom topological descriptors within an aligned point cloud format.<n>For molecular weight, results confirm that AMPTCR encodes physically meaningful data, with a validation R2 of 0.87.<n>In the bacterial inhibition task, AMPTCR enables both classification and direct regression of E. coli inhibition values.
  arXiv  Detail & Related papers  (2025-07-22T04:35:50Z)
• ReDiSC: A Reparameterized Masked Diffusion Model for Scalable Node Classification with Structured Predictions [64.17845687013434]
  We propose ReDiSC, a structured diffusion model for structured node classification.<n>We show that ReDiSC achieves superior or highly competitive performance compared to state-of-the-art GNN, label propagation, and diffusion-based baselines.<n> Notably, ReDiSC scales effectively to large-scale datasets on which previous structured diffusion methods fail due to computational constraints.
  arXiv  Detail & Related papers  (2025-07-19T04:46:53Z)
• Multi-Level Fusion Graph Neural Network for Molecule Property Prediction [8.629821238312621]
  We propose a Multi-Level Fusion Graph Neural Network (MLFGNN) that integrates Graph Attention Networks and a novel Graph Transformer.<n>Experiments on multiple benchmark datasets demonstrate that MLFGNN consistently outperforms state-of-the-art methods in both classification and regression tasks.
  arXiv  Detail & Related papers  (2025-07-04T09:38:19Z)
• Learning Hierarchical Interaction for Accurate Molecular Property Prediction [8.488251667425887]
  Hierarchical Interaction Message Net (HimNet) is a novel deep learning model for predicting ADMET profiles.<n>HimNet achieves the best or near-best performance in most molecular property prediction tasks.<n>We believe HimNet offers an accurate and efficient solution for molecular activity and ADMET property prediction.
  arXiv  Detail & Related papers  (2025-04-28T15:19:28Z)
• RelGNN: Composite Message Passing for Relational Deep Learning [56.48834369525997]
  We introduce RelGNN, a novel GNN framework specifically designed to leverage the unique structural characteristics of the graphs built from relational databases.<n>RelGNN is evaluated on 30 diverse real-world tasks from Relbench (Fey et al., 2024), and achieves state-of-the-art performance on the vast majority tasks, with improvements of up to 25%.
  arXiv  Detail & Related papers  (2025-02-10T18:58:40Z)
• Dumpling GNN: Hybrid GNN Enables Better ADC Payload Activity Prediction Based on Chemical Structure [53.76752789814785]
  DumplingGNN is a hybrid Graph Neural Network architecture specifically designed for predicting ADC payload activity based on chemical structure. We evaluate it on a comprehensive ADC payload dataset focusing on DNA Topoisomerase I inhibitors. It demonstrates exceptional accuracy (91.48%), sensitivity (95.08%), and specificity (97.54%) on our specialized ADC payload dataset.
  arXiv  Detail & Related papers  (2024-09-23T17:11:04Z)
• Synergistic Fusion of Graph and Transformer Features for Enhanced Molecular Property Prediction [0.0]
  We propose a novel approach that combines pre-trained features from GNNs and Transformers. This approach provides a comprehensive molecular representation, capturing both the global molecule structure and the individual atom characteristics. Experimental results on MoleculeNet benchmarks demonstrate superior performance, surpassing previous models in 5 out of 7 classification datasets and 4 out of 6 regression datasets.
  arXiv  Detail & Related papers  (2023-08-25T14:47:46Z)
• 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)
• t-SMILES: A Scalable Fragment-based Molecular Representation Framework for De Novo Molecule Generation [9.116670221263753]
  This study introduces a flexible, fragment-based, multiscale molecular representation framework called t-SMILES. It describes molecules using SMILES-type strings obtained by performing a breadth-first search on a full binary tree formed from a fragmented molecular graph. It significantly outperforms classical SMILES, DeepSMILES, SELFIES and baseline models in goal-directed tasks.
  arXiv  Detail & Related papers  (2023-01-04T21:41:01Z)
• Multi-View Graph Neural Networks for Molecular Property Prediction [67.54644592806876]
  We present Multi-View Graph Neural Network (MV-GNN), a multi-view message passing architecture. In MV-GNN, we introduce a shared self-attentive readout component and disagreement loss to stabilize the training process. We further boost the expressive power of MV-GNN by proposing a cross-dependent message passing scheme.
  arXiv  Detail & Related papers  (2020-05-17T04:46:07Z)
• Assessing Graph-based Deep Learning Models for Predicting Flash Point [52.931492216239995]
  Graph-based deep learning (GBDL) models were implemented in predicting flash point for the first time. Average R2 and Mean Absolute Error (MAE) scores of MPNN are, respectively, 2.3% lower and 2.0 K higher than previous comparable studies.
  arXiv  Detail & Related papers  (2020-02-26T06:10:12Z)

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.