标题： 基于图神经网络的罕见疾病诊断知识图谱稀疏化 显示英文标题

标题： Knowledge Graph Sparsification for GNN-based Rare Disease Diagnosis

摘要： 罕见遗传病的诊断面临重大挑战：患者数据不足、无法获取全基因组测序以及可能致病基因数量庞大。 这些限制导致诊断过程漫长、治疗不当以及关键延误，尤其影响资源有限地区患者，这些地区的诊断工具稀缺。 我们提出了RareNet，一种基于子图的图神经网络，只需患者表型即可识别最可能的致病基因，并提取聚焦的患者子图以进行有针对性的临床研究。 RareNet可以作为独立方法使用，或作为其他候选基因优先级排序方法的预处理或后处理过滤器，持续提升其性能，同时可能实现可解释的洞察。 通过在两个生物医学数据集上的全面评估，我们展示了具有竞争力和鲁棒性的致病基因预测，并在与其他框架集成时表现出显著的性能提升。 通过仅需表型数据，而表型数据在任何临床环境中都易于获取，RareNet使复杂的遗传分析更加普及，特别为缺乏先进基因组基础设施的未受服务群体提供了重要价值。 显示英文摘要

摘要： Rare genetic disease diagnosis faces critical challenges: insufficient patient data, inaccessible full genome sequencing, and the immense number of possible causative genes. These limitations cause prolonged diagnostic journeys, inappropriate treatments, and critical delays, disproportionately affecting patients in resource-limited settings where diagnostic tools are scarce. We propose RareNet, a subgraph-based Graph Neural Network that requires only patient phenotypes to identify the most likely causal gene and retrieve focused patient subgraphs for targeted clinical investigation. RareNet can function as a standalone method or serve as a pre-processing or post-processing filter for other candidate gene prioritization methods, consistently enhancing their performance while potentially enabling explainable insights. Through comprehensive evaluation on two biomedical datasets, we demonstrate competitive and robust causal gene prediction and significant performance gains when integrated with other frameworks. By requiring only phenotypic data, which is readily available in any clinical setting, RareNet democratizes access to sophisticated genetic analysis, offering particular value for underserved populations lacking advanced genomic infrastructure.