Improved Validation of Peptide MS/MS Assignments Using Spectral Intensity Prediction

Abstract

A major limitation in identifying peptides from complex mixtures by shotgun proteomics is the ability of search programs to accurately assign peptide sequences using mass spectrometric fragmentation spectra (MS/MS spectra). Manual analysis is used to assess borderline identifications; however, it is error-prone and time-consuming, and criteria for acceptance or rejection are not well defined. Here we report a Manual Analysis Emulator (MAE) program that evaluates results from search programs by implementing two commonly used criteria: 1) consistency of fragment ion intensities with predicted gas phase chemistry and 2) whether a high proportion of the ion intensity (proportion of ion current (PIC)) in the MS/MS spectra can be derived from the peptide sequence. Recent advances in genome sequencing, MS instrumentation, and chromatographic methods allow high throughput identification of peptide fragmentation spectra (MS/MS spectra)1 from samples as complex as whole cell extracts (1, 2). For very complex samples, the best results are obtained from ion trap mass spectrometers due to their fast scanning rate and ability to rapidly shift between MS and MS/MS modes during data collection. However, when operated for high data collection rate, mass accuracy and resolution are compromised. It was recognized early on that scores from search programs showed poor discrimination between correct and incorrect sequence assignments when using ion trap MS/MS spectral data to search large protein databases (3). Methods have been developed to specify thresholds for acceptance either by searching datasets against an inverted sequence database of similar size to identify false positive thresholds (4) or by statistical analysis of multiple scores and results from normal searches (4, 5). Using methods such as these, limits on search program scores or combinations of scores can be set to yield a low number of false positives (6), but they also will produce large false negative rates (4, 7). This problem is more acute when larger databases are used.To minimize false negatives, investigators often reduce the acceptance threshold to capture more information. Several methods have been developed to filter the resulting false positives based on agreement between sequence composition of the peptides and their behavior on ion exchange or reverse phase chromatography (7, 8), probability of missed cleavages (9), exact mass measurements (8), or differences in scores between the top ranking peptides and lower ranked candidates (10, 11). Methods have also utilized intensity in- 1 The abbreviations used are: MS/MS spectra, fragmentation spectra; ⌬CN, difference between first and second ranked assignments by XCorr; DTA, text file summary of MS/MS spectral information; sDTA, simplified DTA; IntFrag score, proportion of ion current assigned to internal fragment ions;

Unrestrictive Identification of Multiple Post-translational Modifications from Tandem Mass Spectrometry Using an Error-tolerant Algorithm Based on an Extended Sequence Tag Approach

“…A match quality of peptide identification can best be evaluated taking into account various properties of fragmentation patterns in MS/MS spectra (29,30). Our scoring model consists of four scoring components: 1) ion score, 2) standard deviation of mass errors of matched fragment ions, 3) score of explained intensities, and 4) score of explained highest peaks.…”

Tumor Necrosis Factor-α- and Interleukin-1-Induced Cellular Responses: Coupling Proteomic and Genomic Information

“…12) and Mascot (v. 1.9) packages. 19 Peptide identifications were validated using XCorr and MOWSE score filters set in MSPlus. 20 The false discovery rate of peptide identification was determined by searching the Human IPI database with protein sequences that have been reversed.21…”

Charge States of y Ions in the Collision-Induced Dissociation of Doubly Charged Tryptic Peptide Ions

“…Finally, the specific dependences of I(y 2+ )/I(y 1+ ) reported here can find use for spectral prediction [21], scoring [22][23][24][25] and quality control in spectral libraries [13].…”