分子表示学习中的结构层级与几何
Structural Hierarchy and Geometry in Molecular Representation Learning
- Delft University of Technology（代尔夫特理工大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
该研究探讨显式编码Bemis-Murcko骨架并以其监督分子嵌入的效果，对比欧氏与洛伦兹对比目标，发现骨架监督可塑造分子嵌入空间组织，且对分子属性预测有增益，效果因任务和几何类型而异。
AI中文摘要：
分子自监督学习利用化学结构来引导哪些分子嵌入应具有相似性。本研究探讨显式编码分子的Bemis-Murcko骨架并将其用于监督分子嵌入，是否会改变模型的学习内容；还通过对比欧氏和洛伦兹对比目标，测试该效果是否依赖于嵌入几何。在两种增强强度下，骨架监督模型始终根据相同及结构相关的骨架来组织分子，所得嵌入在多项任务的分子属性预测中表现更优，具体增益取决于预测属性。骨架监督对分子组织的影响在洛伦兹目标下更强，但两种几何均未提供一致的整体优势。这些结果表明，明确教授分子与其结构核心的关系可可靠地塑造分子嵌入空间的组织，而该组织的有用程度仍取决于具体任务。
英文摘要：
Molecular self-supervised learning uses chemical structures to guide which molecular embeddings should be similar. We study whether explicitly encoding a molecule's Bemis-Murcko scaffold and using it to supervise the molecular embedding changes what the model learns. We further test whether this effect depends on the embedding geometry by comparing Euclidean and Lorentz contrastive objectives. Across two augmentation strengths, scaffold-supervised models consistently organize molecules according to both identical and structurally related scaffolds. The resulting embeddings also improve molecular property prediction on several tasks, while the exact gains depend on the predicted property. The effect of scaffold supervision on molecular organization is stronger under Lorentz objectives, but neither geometry provides a consistent overall advantage. These results show that explicitly teaching the relation between a molecule and its structural core can reliably shape the organization of molecular embedding space, while the extent of usefulness of this organization remains task dependent.
