Hierarchical Structure-Property Alignment for Data-Efficient Molecular Generation and Editing

Property-constrained molecular generation and editing are crucial in AI-driven drug discovery but remain hindered by two factors: (i) capturing the complex relationships between molecular structures and multiple properties remains challenging, and (ii) the narrow coverage and incomplete annotations of molecular properties weaken the effectiveness of property-based models. To tackle these limitations, we propose HSPAG, a data-efficient framework featuring hierarchical structure–property alignment. By treating SMILES and molecular properties as complementary modalities, the model learns their relationships at atom, substructure, and whole-molecule levels. Moreover, we select representative samples through scaffold clustering and hard samples via an auxiliary variational auto-encoder (VAE), substantially reducing the required pre-training data. In addition, we incorporate a property relevance-aware masking mechanism and diversified perturbation strategies to enhance generation quality under sparse annotations. Experiments demonstrate that HSPAG captures fine-grained structure–property relationships and supports controllable generation under multiple property constraints. Two real-world case studies further validate the editing capabilities of HSPAG.

Authors

• Ziyu Fan Central South University
• Zhijian Huang Central South University
• Yahan Li Central South University
• Xiaowen Hu Central South University
• Siyuan Shen
• Yunliang Wang Xinjiang University
• Zeyu Zhong Central South University
• Shuhong Liu The University of Tokyo
• Shuning Yang Central South University
• Shangqian Wu Central South University
• Min Wu
• Lei Deng

Keywords

No keywords are indexed for this paper.