Analysis of Large LC-MS/MS Datasets Using Amazon Web Services and the Trans-Proteomic Pipeline

摘要

1. Large MS/MS datasets can be cost effectively and expeditiously analyzed with AWS Cloud services. A equivalent analysis on a single 4 core system would take over 12 days. 2. Using iProphet to combine all the results yields a 11% to 40% increase over a single search engine in the number of unique peptides identified with a iProphet probability cutoff at 1% FDR. 3. Costs can be substantially less using EC2 spot pricing ($.22) versus the $ 0.64 reserved instances. 4. There is ample room for improvement in the EC2 provisioning algorithm to decrease the completion time and cost of an analysis