MedRefSNP: A database of medically investigated SNPs

Genetic association studies and linkage analyses using single nucleotide polymorphisms (SNPs) are rapidly increasing in number, and the results are important for evaluating the utility of SNPs in the biomedical sciences. Although many SNP databases have been established, there is no database focusing on published SNPs, where the research results of scientific investigations are available. To enhance the utilization of such SNP data, we have developed the MedRefSNP database (http://www.medclue.com/medrefsnp) to provide integrated information about SNPs collected from the PubMed and OMIM databases. The RefSNP identifiers are automatically identified and are linked to various information sources such as the dbSNP, the HapMap database, the Entrez Gene database, the UCSC genome browser, the CGAP Pathway Searcher, and genetic association databases. And, each SNP is checked to determine whether the PolyDoms, SNPs3D or PolyPhen databases predicts that the SNP affects the phenotype of the protein encoded by the gene carrying the SNP. Also, neighboring SNPs showing strong linkage disequilibrium (LD) with published SNPs are included, using HapMap data. Currently, 36199 unique SNPs (including 31368 neighboring SNPs) collected from 25906 PubMed abstracts and 590 OMIM records are stored along with 2491 human genes related to 466 molecular pathways. The MedRefSNP database will help researchers to review previously investigated results more efficiently, and will expand knowledge by using the genomic and functional contexts of the SNPs.