Search papers, labs, and topics across Lattice.
EvidenceNet is introduced as a framework that leverages LLMs to extract structured, experimentally-grounded findings from full-text biomedical literature and construct disease-specific knowledge graphs. The framework normalizes biomedical entities, scores evidence quality, and connects evidence records through typed semantic relations. The resulting EvidenceNet-HCC and EvidenceNet-CRC datasets demonstrate high component fidelity and improve performance in retrieval-augmented question answering, establishing EvidenceNet as a valuable resource for biomedical reasoning.
LLMs can now construct high-fidelity, disease-specific knowledge graphs from full-text biomedical literature, unlocking evidence-aware reasoning and hypothesis generation.
Biomedical knowledge resources often either preserve evidence as unstructured text or compress it into flat triples that omit study design, provenance, and quantitative support. Here we present EvidenceNet, a framework and dataset for building disease-specific knowledge graphs from full-text biomedical literature. EvidenceNet uses a large language model (LLM)-assisted pipeline to extract experimentally grounded findings as structured evidence nodes, normalize biomedical entities, score evidence quality, and connect evidence records through typed semantic relations. We release two resources: EvidenceNet-HCC with 7,872 evidence records, 10,328 graph nodes, and 49,756 edges, and EvidenceNet-CRC with 6,622 records, 8,795 nodes, and 39,361 edges. Technical validation shows high component fidelity, including 98.3% field-level extraction accuracy, 100.0% high-confidence entity-link accuracy, 87.5% fusion integrity, and 90.0% semantic relation-type accuracy. In downstream evaluation, EvidenceNet improves internal and external retrieval-augmented question answering and retains structural signal for future link prediction and target prioritization. These results establish EvidenceNet as a disease-specific resource for evidence-aware biomedical reasoning and hypothesis generation.
