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

TL;DR

HSPAG is a hierarchical, data-efficient framework that improves property-constrained molecular generation and editing by learning structure-property relationships across multiple levels and reducing data requirements.

Contribution

The paper introduces HSPAG, a novel hierarchical structure-property alignment model that enhances data efficiency and controllable molecular generation with sparse annotations.

Findings

HSPAG effectively captures detailed structure-property relationships.

The model supports controllable generation under multiple property constraints.

Case studies demonstrate successful molecular editing capabilities.

Abstract

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.