The National Institutes of Health (NIH) has launched Linked Discoveries, an experimental tool designed to help researchers better understand how individual scientific findings connect to the broader body of biomedical evidence. The platform allows users to explore networks of related publications directly from articles indexed in PubMed, including studies that replicate or build upon previous research.
Developed by the National Library of Medicine (NLM), which manages PubMed, Linked Discoveries represents an early outcome of NIH's agency-wide initiative to strengthen replication and reproducibility in federally funded research. The tool is intended to address a longstanding challenge in scientific research: determining how individual findings fit within the wider evidence landscape.
At launch, Linked Discoveries includes more than 29 million publications from PubMed. The platform provides researchers with a connected view of related studies, helping them identify relevant evidence that may otherwise be scattered across multiple sources. The initiative also aims to reduce reliance on traditional literature searches, which can favor highly cited or already prominent publications.
The platform uses an AI-informed approach to identify related publications and visualize their connections through interactive graph and timeline formats. Users can refine searches based on specific conditions, genes, and chemicals while accessing information such as citation relationships, review articles, retractions, and NIH-funded studies.
By presenting these connections, Linked Discoveries provides additional context that can help researchers evaluate existing evidence, interpret previous findings, and design future studies. The platform does not assess the quality of research or determine whether findings have been successfully replicated. Instead, it organizes relationships among publications, allowing researchers to make their own evidence-based assessments.
The September 24 release marks the first public version of Linked Discoveries. NLM plans to continue refining the platform based on user feedback submitted through the tool. Future enhancements are expected to support NIH's broader efforts to improve research replication and reproducibility while maintaining researchers' responsibility for scientific evaluation and interpretation.
Click here to read the original press release.