FOR IMMEDIATE RELEASE
June 10, 2020 ANN ARBOR, MI
Genomenon’s latest release of the Mastermind® Genomic Search Engine includes enhanced search for intronic and non-coding variants across the medical literature. Mastermind users now receive more specific results when searching the comprehensive database of genomic evidence.
Intronic and non-coding variants are notoriously difficult for geneticists and variant scientists to find with search tools like Google or Google Scholar. Unlike missense variants, which sit in the coding region of a gene and can be described by the effect on the gene’s transcribed protein, intronic variants are much more difficult to search for because they have a less direct and well-understood biological effect.
With the latest release of the Mastermind Genomic Search Engine, Genomenon increased the specificity of the intronic and non-coding variant search results by prioritizing nucleotide-specific descriptions within the search results. This provides the user with optimal specificity while maintaining maximal sensitivity. By searching at the nucleotide level for non-coding variants, users can now see the most specific results in the top articles returned by Mastermind. Broader sensitivity is delivered with similar intronic variants in the same region further down the search results.
Genomenon has always striven to balance sensitivity and specificity in search, which has consistently yielded better results than other methods. One example is a user of Mastermind who was searching for an intronic deletion. Before Mastermind, they had never found any evidence citing this variant for a cancer patient. However, the sensitivity of the Mastermind search returned a previously undiscovered article. This new approach to prioritization will further optimize these types of searches.
This new feature furthers Genomenon’s position as the leading genomic search engine in the market – making it easier for geneticists, variant scientists, and researchers to comprehensively search across the scientific literature for every mention of a patient’s variant.
The Real-world Evidence to validate a drug target, identify trial-eligible patients, or change a diagnosis already exists. It is buried in 39 million biomedical articles, locked behind paywalls and supplemental files most researchers never find.
Genomenon closes that gap. Fit-for-purpose AI-powered search reads 11.2 million full-text papers and 3.7 million supplemental datasets. Eighty expert scientific curators validate every finding. The result is structured, traceable, regulatory-grade Real-World Evidence at the genetic variant and patient level. Loxo@Lilly used Genomenon to add 73 variants to the RET label. In a head-to-head, Genomenon identified 83% more rare disease patients than ChatGPT plus OpenEvidence.
250+ diagnostic labs and 75 biopharma programs rely on Genomenon as the evidence layer behind precision medicine.
