Learning Molecular Representations from Cellular Phenotypes with Structure Preservation

通过细胞表型学习保持结构特征的分子表征

Phenotypic drug discovery enables the discovery of functional relationships between molecular structures and cellular responses. However, existing multimodal representation learning methods often optimize cross-modal alignment without considering the intrinsic organization of chemical space, resulting in distorted molecular representations and loss of structural information.

表型药物发现能够揭示分子结构与细胞反应之间的功能关系。然而，现有的多模态表征学习方法往往在优化跨模态对齐时，忽略了化学空间的内在组织结构，导致分子表征扭曲以及结构信息的丢失。

We propose \textbf{PhenMol}, a structure-preserving framework for phenotype-aware molecular representation learning. PhenMol disentangles molecular and cellular representations into shared and private components, enabling phenotype-guided alignment while preserving chemical structures through a dedicated molecular branch. This design integrates cellular phenotype information without disrupting molecular neighborhood organization.

我们提出了 PhenMol，这是一个用于表型感知分子表征学习的结构保持框架。PhenMol 将分子和细胞表征解耦为共享组件和私有组件，在实现表型引导对齐的同时，通过专门的分子分支保留化学结构。这种设计在整合细胞表型信息的同时，不会破坏分子邻域的组织结构。

Experiments on approximately $3.04 \times 10^{4}$ molecule—cell morphology pairs demonstrate that PhenMol improves molecular property prediction across 270 bioactivity tasks, molecule—phenotype retrieval, and clinical trial outcome prediction. Moreover, ECFP4-based structural analysis shows that PhenMol better preserves molecular neighborhoods and reduces embedding distortion compared with existing multimodal alignment methods.

在约 $3.04 \times 10^{4}$ 对分子-细胞形态数据上的实验表明，PhenMol 在 270 项生物活性任务、分子-表型检索以及临床试验结果预测中，均提升了分子属性预测的性能。此外，基于 ECFP4 的结构分析显示，与现有的多模态对齐方法相比，PhenMol 能更好地保留分子邻域并减少嵌入扭曲。

These results highlight the importance of structure-aware constraints in multimodal representation learning and provide an effective approach for integrating cellular phenotypes with chemical knowledge for drug discovery.

这些结果凸显了结构感知约束在多模态表征学习中的重要性，并为在药物发现中整合细胞表型与化学知识提供了一种有效的方法。