The effects of mass accuracy, data acquisition speed, and search algorithm choice on peptide identification rates in phosphoproteomics

Proteomic analyses via tandem mass spectrometry have been greatly enhanced by the recent development of fast, highly accurate instrumentation. However, successful application of these developments to high-throughput experiments requires careful optimization of many variables which adversely affect each other, such as mass accuracy and data collection speed. We examined the performance of three shotgun-style acquisition methods ranging in their data collection speed and use of mass accuracy in identifying proteins from yeast-derived complex peptide and phosphopeptide-enriched mixtures. We find that the combination of highly accurate precursor masses generated from one survey scan in the FT-ICR cell, coupled with ten data-dependent tandem MS scans in a lower-resolution linear ion trap, provides more identifications in both mixtures than the other examined methods. For phosphopeptide identifications in particular, this method identified over twice as many unique phosphopeptides as the second-ranked, lower-resolution method from triplicate 90-min analyses (744 +/- 50 vs. 308 +/- 50, respectively). We also examined the performance of four popular peptide assignment algorithms (Mascot, Sequest, OMSSA, and Tandem) in analyzing the results from both high-and low-resolution data. When compared in the context of a false positive rate of approximately 1%, the performance differences between algorithms were much larger for phosphopeptide analyses than for an unenriched, complex mixture. Based upon these findings, acquisition speed, mass accuracy, and the choice of assignment algorithm all largely affect the number of peptides and proteins identified in high-throughput studies.