Literature mining for rare disease phenotype genotype associations
Year: 2020
Type: other
Abstract: Although individually uncommon, collectively, rare diseases affect 6 8% of the world’s population. Diagnosing rare diseases through phenotypes is a difficult task because symptoms overlap based solely on phenotype there may be multiple causes for a single disease and a single cause may be associated with multiple diseases. Genomic investigations are leading to precise molecular level characterization allowing for a systematic discovery of therapies that either target a specific disease or, more broadly, multiple related diseases. Currently, text mining techniques have lower accuracy and more coverage gaps than manual curation, but the potential for higher productivity. Recently, there has been some progress in developing text mining applications to tackle this enormous problem. However, there is still a need to assess these text mining applications and integrate the findings with data. Towards these ends, we manually created a rare disease variant data set which can be used to test and refine a text mining algorithm. We developed a manual curation workflow which incorporated searching for genetic variants on PubMed, listing symptoms and phenotypes associated with each variant, converting cDNA or protein to standardized genetic notation, obtaining annotations from public data sources, and presenting the details in an online interface for future release. Our project aims include summation of pathogenic variant frequency in populations to estimate birth prevalence for each rare disease. We created test data sets of known accuracy, coverage, and genotype phenotype associations that can be used to validate text mining approaches. Enhanced text mining will significantly decrease the amount of time necessary to gather data to molecularly characterize a rare disease and render it possible to mine rare disease phenotype genotype associations for the more than 7,000 rare disease genes in a timely fashion.
Authors: Erica Lyons, Daniel Watson, Jack Collins, Gregory Tawa, Sharie Haugabook, Elizabeth Ottinger, Uma Mudunuri
Institution: Frederick National Laboratory for Cancer Research
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: gold
OpenAlex ID: W3087458764