Skip to content
Open access

Reaction-aware hypergraph learning of chemical representations with LARK

Oct 2026 · bioRxiv · 0 citations · 22 references
Biology

TL;DR

LARK is introduced, a unified framework for learning atom–molecule–reaction knowledge from chemical transformations that learns chemically informative representations of substructures involved in molecular transformations and establishes chemical transformations as a source of structured supervision for representation learning across molecular and reaction scales.

Abstract

Molecular pretraining offers a route to learning transferable chemical representations from unlabelled data. However, existing approaches pretrained on isolated molecular structures struggle to generalize to reaction-specific tasks because their pretraining objectives provide limited information about chemical transformations. A key challenge is to integrate this reaction information into chemical representations. Here we introduce LARK, a unified framework for learning atom–molecule–reaction knowledge from chemical transformations. Using approximately 0.6 million reactions encompassing 0.88 million unique molecular components, LARK combines bidirectional bond–electron reconstruction with structural and geometric supervision through a role-aware hierarchical hypergraph Transformer that links local atom representations to reaction context. The learned representations transfer to both reaction prediction and single-molecule property prediction, with strong performance across 42 tasks covering reaction outcomes, physicochemical properties, toxicity, metabolism and pharmacokinetics. Reactioncentre analyses show that LARK identifies where bonds break and form, while targeted interventions reveal that the learned atom representations contribute to reconstructing these changes. These findings indicate that LARK learns chemically informative representations of substructures involved in molecular transformations. Together, the results establish chemical transformations as a source of structured supervision for representation learning across molecular and reaction scales.

Read PDF

Similar papers

#artificial intelligence Preprint Sep 2026

Learning Transferable Reaction Mechanisms from Visual Chemical Knowledge

This work introduces MechaVLM, a visual framework that combines transferable chemical representations with external mechanistic knowledge that establishes strong zero-shot mechanism prediction and generalizes beyond mechanism prediction to atom mapping and reaction center prediction.

Yu-Jian Yuan, Jia-Xin Xu, Xin-Ying Cai et al. · 0 citations
Open access Oct 2026

GRASP: Graph Representation Learning with Assay Supervision for Molecular Properties

Molecular structures are abundant, but measured properties are scarce and scattered across assays that each cover different compounds. Most molecular pretraining pipelines use only the structures, moving directly from self-supervision to endpoint-specific fine-tuning and leaving large, sparsely labelled bioactivity dat...

S. P. Srivastava, R. Gorantla, Sharath Krishna Chundru et al. · 2 citations
#artificial intelligence Preprint Oct 2026

Elucidating the Space of Enzymatic Reaction: A Unified Benchmark and Pretrained Model

Existing reaction models primarily learn molecular transformations, whereas enzy- matic reactions depend jointly on molecular structure and catalytic function. We formulate this problem as learning an enzymatic reaction space linking reactants, products, and Enzyme Commission (EC) annotations. To characterize this spac...

Yu-Tong Hu, Tian-Ming Huang, Yan-Bo Zhao et al. · 0 citations
Open access Sep 2026

Geo-Hete-HyperGNN as a prior-free molecular hypergraph representation learning strategy

Molecular representation learning is a cornerstone of AI-driven chemical discovery. However, most molecular hypergraph models define hyperedges using expert-crafted rules and often overlook three-dimensional molecular geometry. Motivated by the close correspondence between molecular structure and spectroscopy, includ...

Wei Zhou, Ying-Zi Peng, Mengtao Mo et al. · 0 citations
Open access Sep 2026

ActiveFusion: Fused Representations Improve Active Learning for Molecular Property Prediction

This work introduces ActiveFusion, a framework for integrating heterogeneous molecular representations within active learning workflows for molecular property prediction and offers practical guidance for designing active learning pipelines that balance prediction accuracy with chemical space exploration.

Nelson Evbarunegbe, Shiyun Wa, Luke Taylor et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.