PepArML Meta-Search Engine
The PepArML Meta-Search Engine provides a unified interface for running tandem mass spectrometry searches with Mascot, Tandem, OMSSA, KScore (Tandem with the K-Score scoring plugin), SScore, Myrimatch, and InsPecT on a heterogeneous grid of computers locally on the Edwards Lab cluster and remotely using Amazon Web Services or NSF's TeraGrid infrastructure.
Worker computers contact the PepArML meta-search engine server to download spectra, sequence databases, and search parameters; search the spectra against the appropriate sequence database using the required search engine; and upload the results back to the search server. All data is transferred in a secure manner, with the entire worker-scheduler communication channel encrypted using the https protocol, and worker requests authenticated by association with users. Computational resources supplied by users are applied only to their search jobs, while PepArML resources (approx 80 CPUs) are shared equitably across users' search jobs.
The PepArML Meta-Search Engine is free - users can register immediately on-line. Compute resources are added or removed dynamically with demand.
Spectra are uploaded for analysis under the Spectra tab, and new searches configured under the Search tab. Search jobs can be monitored using the Queue, Running, Error, and Done tabs. Search results can be found under the Results tab, and results selected for combining, with parameters determined under the Combine tab.
A variety of publicly accessible spectra and results are available under the Spectra and Results tabs. In particular the results for Peptide Atlas dataset A8_IP and Sashimi dataset S17, mentioned in various PepArML talks, papers, and posters can be found here. See also the tutorial and worked example at the Edwards Lab blog.
Please contact Nathan Edwards (nje5 at georgetown.edu) for more information or assistance.