The answers to many questions are sitting in the primary scientific literature. The challenge is efficiently reading through millions of documents to find them.
GeneIndexer, a cutting-edge semantic engine, was designed to efficiently reveal known and unknown genetic relationships using all relevant genomic literature. GeneIndexer’s technology significantly compresses the time needed for discovery. This capability allows Computable Genomix to quickly assemble high-value genetic targets for biomarker discovery and companion diagnostic development.
The Secret Sauce
GeneIndexer uses a powerful text-mining approach to automatically identify and model is known (explicit) and unknown (implicit) gene and SNP relationships from MEDLINE abstracts.
GeneIndexer deduces gene-to-gene and gene-to-keyword relationships. It extracts features and relationships from the scientific literature that are not easily made or even possible by humans. With GeneIndexer, Computable Genomix can rapidly mine the biomedical literature and make mechanistic or functional predictions that were not previously possible.
A project involving Cisplatin-induced hearing loss provides an excellent example of GeneIndexer’s ability to identify unknown relationships. The queries “Cisplatin” and “hearing loss” were used to identify genes that are functionally related to the queries. In addition to numerous known candidates, GeneIndexer identified a few genes whose abstracts made no mention of Cisplatin, chemotherapy, hearing loss, or deafness. One of these genes was, however, implicated in a hereditary disease known to cause deafness-onset in childhood and to be a known regulator of mitochondrial activity.