One of the key pieces of the drug discovery puzzle involves determining the clinical relevance of the target. Over the past few years big pharma have been pulling back from this critical part of the drug discovery process and ceding this territory to biotech startups and academia. Until recently, the only major database that could provide this type of information was OMIM (Johns Hopkins Online Mendelian Inheritance In Man). However, the focus at OMIM was more on providing a wide-angle view of the genes involved in particular diseases.
Moreover, if you wanted to understand a disease and all potential targets for that disease, the most you could expect from an OMIM record was the Reader’s Digest perspective. You have only to compare the OMIM record for pancreatic cancer, with the NCBI’s eBook on the same subject to get an idea of the difference in the depth of coverage of the subject.
Recently, the NCBI began work on the ClinVar project. The purpose of ClinVar is to “provide a freely accessible, public archive of reports of the relationships among human variations and phenotypes along with supporting evidence”. This YouTube video gives you an overview of ClinVar. The data in ClinVar includes records found in OMIM, but also data from dbSNP and other data sources.
One caveat about ClinVar, the database is still a work-in-progress. A simple search for “pancreatic cancer” yields hits for hereditary pancreatitis, whereas a search for “pancreatic adenocarcinoma” yields a much more targeted set of hits. The results are also incomplete, since you won’t find any references for the KRASG12D mutation found in the majority of pancreatic cancers, nor will you find reference to mutations in TP53 (found in approximately 50% of patients), or BRCA2 mutations.