Disease-Associated Genes Extraction from Literature Database

Databases such as MEDLINE, OMIM, RefSeq and LocusLink serve as important information sources for biomedical studies. A search of several databases can yield vast quantities of information, so much in fact that the searcher may struggle to pinpoint the important information. This has highlighted the desirability of a specific tool to facilitate efficient database searching and extraction of information on undefined relationships (e.g., the relationship between a disease and certain genes). At present, tools such as ArrowSmith [3], BITOLA [1] and G2D [2] are available for this purpose. All of these tools utilize co-occurrence relation and other semantic and grammatical characterizations to unearth the required information. ArrowSmith and BITOLA are interactive tools designed to search undefined relationships between terms which are not co-occurrent in a single published article. G2D provides a scoring system for the possible functional relations of human genes to genetically inherited diseases that have been mapped onto chromosomal regions without assignment of a particular gene. These tools suggest the linkage between a given disease and certain genes by tracing chains of terms. However, since only one or two intermediate terms are presented, the leap or variance of meaning between terms is too great to enable precise relation of a disease to genes. If such leaps can be minimized, we will be able to glean more information more easily.