Using subspectral interval matching to make novel identifications of ETD tandem mass spectra

Tuesday, October 25, 2011 — Poster Session II
Noon – 2:00 p.m.	Natcher Conference Center	NLM	PROTEOM-7

Authors

• V Sridhara
• DL Bai
• A Chi
• J Shabanowitz
• DF Hunt
• SH Bryant
• LY Geer

Abstract

Many existing search methods for tandem mass spectrometry are designed to match entire spectra containing canonical ion series. For example, de novo and sequence tag methods often require consecutive ion runs. Sequence search methods attempt to identify high mass ions which can be combinatorically difficult to match due to modifications. Given the significant number of unidentified spectra in a typical experiment, algorithms that circumvent these constraints may yield novel identifications. The OMSSA-SIM algorithm matches short mass intervals between intense peaks to those generated from sequence library peptides. The algorithm was written in c++. ETD MS/MS spectra of yeast phosphopeptides are used in this study. In a first pass of the algorithm, the experimental mass intervals from the spectra were matched to the theoretical mass intervals from the peptides. High scoring matches were retained, and, in second pass of the algorithm, used as anchors and the hits extended in both directions of the sequence until a maximally scoring match was obtained. The algorithm was run using both target and randomly shuffled yeast sequence libraries and compared to results from the original OMSSA. From this preliminary dataset, OMSSA-SIM yields 50% additional hits to our test spectra with a FDR of 9%.

back to top