Research Papers 论文研究 12h ago Updated 1h ago 更新于 1小时前 35

HypoKG: Evidence-Disciplined Biomedical Hypothesis Generation Beyond Endpoint Knowledge HypoKG: Evidence-Disciplined Biomedical Hypothesis Generation Beyond Endpoint Knowledge

HypoKG integrates KEGG, Rhea, and UniProt into a unified biochemical knowledge graph to benchmark LLM-based biomedical hypothesis generation Six LLMs generated 13,200 hypotheses across four conditions varying the biological information provided (source enzyme only, full path, endpoints only) Models given only source and disease endpoints produced the most compelling hypotheses but were least evidence-grounded Full biological path information led to "evidence-disciplined reasoning" — hypotheses m 提出HypoKG框架,整合KEGG、Rhea、UniProt三大生物数据库构建统一生化知识图谱,用于评估LLM在生物医学假设生成中的证据推理能力 构建包含550条酶源到罕见病终点路径的基准测试,生成13,200个假设,采用专家制定的五标准评分体系(1-5分) 发现LLM在仅获知源酶和疾病终点时能生成高分但证据基础薄弱的假设,而提供完整生物路径时产生更符合已知机制关系的"证据纪律推理" 通过打乱中间路径步骤的对照实验证实模型确实利用路径结构进行推理(delta = -0.793, p < 0.001) 知识图谱在假设生成中发挥双重作用:识别文献中缺失的生物终点对,并通过机制路径引导LLM推理过程

50
Hot 热度
50
Quality 质量
50
Impact 影响力

Analysis 深度分析

TL;DR

  • HypoKG integrates KEGG, Rhea, and UniProt into a unified biochemical knowledge graph to benchmark LLM-based biomedical hypothesis generation
  • Six LLMs generated 13,200 hypotheses across four conditions varying the biological information provided (source enzyme only, full path, endpoints only)
  • Models given only source and disease endpoints produced the most compelling hypotheses but were least evidence-grounded
  • Full biological path information led to "evidence-disciplined reasoning" — hypotheses more consistent with known mechanistic relationships
  • Shuffling intermediate path steps while keeping endpoints fixed caused a significant drop in evidence grounding (delta = -0.793, p < 0.001), confirming models genuinely used path structure

Why It Matters

This research directly addresses a critical concern in AI for science: whether LLMs are truly reasoning from evidence or merely generating plausible-sounding outputs. For biomedical researchers and AI practitioners, it demonstrates that knowledge graphs serve a dual purpose — identifying novel endpoint pairs and structuring the reasoning process itself. The findings have implications for how we design AI-assisted drug discovery and hypothesis generation pipelines.

Technical Details

  • Knowledge Graph Construction: Unified three major biological databases (KEGG, Rhea, UniProt) into a single biochemical knowledge graph
  • Benchmark Design: 550 paths connecting enzyme sources to rare disease endpoints, generating 13,200 hypotheses from six LLMs under four experimental conditions
  • Evaluation Framework: Expert-derived five-criterion rubric scored on a 1-5 scale per criterion, assessing hypothesis quality and evidence grounding
  • Ablation Study: Shuffled intermediate path steps while preserving endpoints to isolate the effect of path structure on reasoning quality
  • Key Metric: Evidence grounding score showed statistically significant degradation (delta = -0.793, p < 0.001) when path structure was disrupted

Industry Insight

  • Knowledge graphs should be viewed as reasoning scaffolds, not just knowledge repositories — their structural information actively shapes LLM reasoning quality in scientific domains
  • There is a trade-off between hypothesis creativity and evidence grounding: minimal information yields more compelling but less reliable hypotheses, while full mechanistic paths produce more disciplined but potentially less novel outputs
  • For AI-driven drug discovery pipelines, combining both approaches — using endpoint-only generation for ideation and path-guided generation for validation — could optimize the balance between novelty and reliability

TL;DR

  • 提出HypoKG框架,整合KEGG、Rhea、UniProt三大生物数据库构建统一生化知识图谱,用于评估LLM在生物医学假设生成中的证据推理能力
  • 构建包含550条酶源到罕见病终点路径的基准测试,生成13,200个假设,采用专家制定的五标准评分体系(1-5分)
  • 发现LLM在仅获知源酶和疾病终点时能生成高分但证据基础薄弱的假设,而提供完整生物路径时产生更符合已知机制关系的"证据纪律推理"
  • 通过打乱中间路径步骤的对照实验证实模型确实利用路径结构进行推理(delta = -0.793, p < 0.001)
  • 知识图谱在假设生成中发挥双重作用:识别文献中缺失的生物终点对,并通过机制路径引导LLM推理过程

为什么值得看

本文为LLM在生物医学领域的假设生成能力提供了严谨的实证评估,揭示了"听起来合理"与"证据 grounded"之间的关键差异。研究结果对构建可信AI辅助科学发现系统具有重要指导意义,特别是在知识图谱与LLM结合的应用场景中。

技术解析

  • 知识图谱构建:整合KEGG(代谢通路)、Rhea(生化反应)和UniProt(蛋白质信息)三个权威生物数据库,构建统一的生化知识图谱,覆盖酶-反应-疾病的多层次关系。
  • 基准测试设计:构建550条连接酶源到罕见病终点的路径,在四种实验条件下(仅源酶、完整生物路径、源酶+疾病终点、源酶+疾病终点+打乱路径)由6个LLM生成共13,200个假设。
  • 评估方法:采用专家制定的五标准评分表(1-5分制),从证据基础、机制一致性等维度对假设质量进行量化评估。
  • 对照实验:通过打乱中间路径步骤而保持端点不变的消融实验,验证模型是否真正利用路径结构进行推理,结果显示证据基础显著下降(delta = -0.793, p < 0.001)。

行业启示

  • 知识图谱+LLM的协同价值:知识图谱不仅提供事实性知识,更能通过结构化路径引导LLM的推理过程,提升生成内容的科学可信度,为"AI for Science"提供可复用的方法论。
  • 评估框架的重要性:在生物医学等高风险领域,仅凭假设的"说服力"不足以评估LLM能力,需建立基于专家知识的严谨评估体系,区分"听起来合理"与"证据 grounded"。
  • 科学发现AI系统的构建方向:未来AI辅助科学发现系统应注重知识图谱的结构化推理能力,而非仅依赖LLM的生成能力,特别是在假设生成、机制推断等需要严格证据支撑的任务中。

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。