NIH Research Festival
Rare diseases affect millions globally, but information and resources are often scarce. NCATS has formed a collaboration with the Frederick National Lab to create a user-friendly platform for rare disease information. The main objective for the project, named RARe-SourceTM, is to provide an innovative application and searchable interface for data mining, by integrating various bioinformatics databases and enabling users to navigate the wealth of information quickly and efficiently.
Biomedical literature is largely unstructured, extracting relevant details is still challenging, and mining for contextual information such as variant pathogenicity, clinical and phenotypic details, and their relations to the genome are not yet resolved.
Developing a scalable and agnostic literature mining workflow is especially important for the rare disease field. Diagnoses of the more than 3000 rare diseases with a known genetic etiology is still challenging and many causative genetic variants are still languishing in published articles without easy access.
We initiated development on the literature AI feature in RARe-SourceTM with the goal of mining relevant rare disease details and making the information accessible to researchers, clinicians, and patient caregivers. Customized implementation of novel natural language processing models will help assess the fit of the models to rare disease research.
Development on the literature AI feature is split into several phases to enable access to results at multiple stages of implementation. The first phase of the project was successfully completed and the results on finding rare disease related articles along with information on associated genes is available through RARe-SourceTM.
Scientific Focus Area: Computational Biology
This page was last updated on Monday, September 25, 2023