通过拓扑感知与知识蒸馏，实现多层生物网络零样本交互预测。
Distilling and Adapting: A Topology-Aware Framework for Zero-Shot Interaction Prediction in Multiplex Biological Networks
- 利用领域基础模型生成增强嵌入，融合结构与序列信息。
- 拓扑感知图分词器捕捉多层连通性，提升高阶关系建模能力。
- 基于对比学习的师生蒸馏策略，支持无历史信息实体的零样本预测。
多层生物网络（MBNs）通过表征实体间的多种相互作用类型，在理解复杂生物系统中具有关键作用。然而，现有方法常难以有效建模多层特性，难以整合结构与序列信息，并在面对无历史邻域信息的新实体时，难以实现零样本交互预测。为此，我们提出一种新型框架，通过上下文感知表示学习与知识蒸馏，实现对MBNs的零样本交互预测。该方法利用领域专用基础模型生成丰富嵌入，引入拓扑感知图分词器以捕捉多层特性和高阶连通性，并采用对比学习对齐跨模态嵌入。进一步通过教师-学生蒸馏策略，实现稳健的零样本泛化。实验结果表明，该框架在MBNs交互预测任务上优于现有最先进方法，为探索多种生物互作及推动个性化治疗提供了有力工具。
原文摘要 · Abstract (English)
Multiplex Biological Networks (MBNs), which represent multiple interaction types between entities, are crucial for understanding complex biological systems. Yet, existing methods often inadequately model multiplexity, struggle to integrate structural and sequence information, and face difficulties in zero-shot prediction for unseen entities with no prior neighbourhood information. To address these limitations, we propose a novel framework for zero-shot interaction prediction in MBNs by leveraging context-aware representation learning and knowledge distillation. Our approach leverages domain-specific foundation models to generate enriched embeddings, introduces a topology-aware graph tokenizer to capture multiplexity and higher-order connectivity, and employs contrastive learning to align embeddings across modalities. A teacher-student distillation strategy further enables robust zero-shot generalization. Experimental results demonstrate that our framework outperforms state-of-the-art methods in interaction prediction for MBNs, providing a powerful tool for exploring various biological interactions and advancing personalized therapeutics.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。