Bringing flexibility to virtual screening for enzymatic inhibitors on the grid

Virtual screening has become an important part of the drug discovery process. Grid computing facilitates this process by providing shared computational resources across different international institutions to run computationally intensive, scientific applications without the need for a centralized supercomputer. This study designed and implemented a flexible, scalable platform to perform a large virtual screening experiment on the PRAGMA grid testbed using the molecular docking simulation software DOCK 5.4. Using Opal OP to wrap DOCK as a grid service and PERL for data manipulation purposes, the ldquodruglikerdquo subset of the ZINC database, which contains 2,066,906 compounds, was successfully screened against the catalytic site of a protein tyrosine phosphatase. The screening required 11.56 days laboratory time and utilized 200 processors over 7 clusters. A ranked list of the best binding compounds to the phosphatase was generated and is currently being tested in biological applications for their efficacy and specificity.