NLM’s LitVar 2.0 Provides More Accurate & Comprehensive Way to Search for Genetic Variants in Biomedical Literature
A team of researchers from the National Library of Medicine (NLM) at the National Institutes of Health (NIH) and the Broad Institute developed an improved web-based system, LitVar 2.0, to help researchers and clinicians search for genetic variants and related information in the biomedical literature. A paper about this powerful LitVar 2.0 system was published in Nature Genetics.
The identification, curation, and interpretation of genomic variants plays a key role in the diagnosis and clinical care of people with genetic diseases and in related research. Finding and keeping up with the latest variant related information is critical for researchers, clinical laboratories, health care professionals, and database curators. However, expert curated variant information is often incomplete due to the sheer volume and rapid growth of medical literature and the high cost of manual curation.
LitVar 2.0 represents a significant advancement in variant information retrieval through its use of state-of-the-art artificial intelligence (AI) to overcome various limitations observed in existing keyword-based search engines and its predecessor. This allows LitVar 2.0 to provide the most accurate and comprehensive results to date.
"Unlike traditional literature search engines, LitVar 2.0 provides a unified search over abstracts, full text, and supplementary data, as well as precise variant recognition through AI-based disambiguation," said Zhiyong Lu, PhD, FACMI, Senior Investigator in the Computational Biology Branch of NLM’s Intramural Research Program. "LitVar 2.0 provides the scientific community with a powerful tool to track the latest variant related publications, which is of critical importance for genomic research and precision medicine."
In addition to improved variant recognition accuracy and the inclusion of variant information from article supplementary data, LitVar 2.0 features more powerful search capabilities, and a redesigned user interface for more convenient results navigation compared with its original system. As of May 2023, LitVar 2.0 indexes approximately 15million unique variants found in the entire biomedical literature and is linked in resources such as dbSNP, ClinVar, and ClinGen.
To learn more about LitVar 2.0 and its features, watch this video.
This research was supported by the NIH Intramural Research Program, National Library of Medicine, National Human Genome Research Institute (grant #U24HG006834).