semantic-gene

Candidate gene prediction using semantic mapping of gene biological annotation data. STIG (Semantic Text Indexing of Genes) ranks genes based on semantic similarity of annotations to text descriptions of Mendelian or complex traits. A manuscript describing STIG is in preparation. A previous approach, CAESAR (CAndidatE Search And Rank), uses text and data mining to rank genes for relevance to an input trait of interest using a variety of gene functional annotations. A more in depth description of CAESAR can be found here: Gaulton KJ et al. A computational system to select candidate genes for complex human traits. Bioinformatics. 2007 May 1;23(9):1132-40. Epub 2007 Jan 19. Contact Kyle Gaulton, kgaulton@gmail.com, with questions or comments.

caesar
README
vector
xcaesar
db
bin