Improving Sensitivity by Probabilistically Combining Results from Multiple MS/MS Search Methodologies

@article{citeulike:2320290, abstract = {Abstract: Database-searching programs generally identify only a fraction of the spectra acquired in a standard LC/MS/MS study of digested proteins. Subtle variations in database-searching algorithms for assigning peptides to MS/MS spectra have been known to provide different identification results. To leverage this variation, a probabilistic framework is developed for combining the results of multiple search engines. The scores for each search engine are first independently converted into peptide probabilities. These probabilities can then be readily combined across search engines using Bayesian rules and the expectation maximization learning algorithm. A significant gain in the number of peptides identified with high confidence with each additional search engine is demonstrated using several data sets of increasing complexity, from a control protein mixture to a human plasma sample, searched using SEQUEST, Mascot, and X! Tandem database-searching programs. The increased rate of peptide assignments also translates into a substantially larger number of protein identifications in LC/MS/MS studies compared to a typical analysis using a single database-search tool.}, address = {Proteome Software Inc., 1340 S.W. Bertha Boulevard, Suite 201, Portland, Oregon 97219-2039, and Department of Pathology and Center for Computational Medicine and Biology, University of Michigan, 1301 Catherine Road, Ann Arbor, Michigan 48109-0602}, author = {Searle, Brian C. and Turner, Mark and Nesvizhskii, Alexey I. }, citeulike-article-id = {2320290}, doi = {10.1021/pr070540w}, journal = {J. Proteome Res.}, keywords = {machine-learning, peptide-identification, peptide-identification-combiners, proteomics, tandem-mass-spectrometry}, month = {January}, number = {1}, pages = {245--253}, posted-at = {2008-02-01 18:43:14}, priority = {2}, title = {Improving Sensitivity by Probabilistically Combining Results from Multiple MS/MS Search Methodologies}, url = {http://dx.doi.org/10.1021/pr070540w}, volume = {7}, year = {2008} }

TY - JOUR ID - citeulike:2320290 L3 - citeulike-article-id:2320290 TI - Improving Sensitivity by Probabilistically Combining Results from Multiple MS/MS Search Methodologies JF - J. Proteome Res. VL - 7 IS - 1 SP - 245 EP - 253 AD - Proteome Software Inc., 1340 S.W. Bertha Boulevard, Suite 201, Portland, Oregon 97219-2039, and Department of Pathology and Center for Computational Medicine and Biology, University of Michigan, 1301 Catherine Road, Ann Arbor, Michigan 48109-0602 N2 - Abstract: Database-searching programs generally identify only a fraction of the spectra acquired in a standard LC/MS/MS study of digested proteins. Subtle variations in database-searching algorithms for assigning peptides to MS/MS spectra have been known to provide different identification results. To leverage this variation, a probabilistic framework is developed for combining the results of multiple search engines. The scores for each search engine are first independently converted into peptide probabilities. These probabilities can then be readily combined across search engines using Bayesian rules and the expectation maximization learning algorithm. A significant gain in the number of peptides identified with high confidence with each additional search engine is demonstrated using several data sets of increasing complexity, from a control protein mixture to a human plasma sample, searched using SEQUEST, Mascot, and X! Tandem database-searching programs. The increased rate of peptide assignments also translates into a substantially larger number of protein identifications in LC/MS/MS studies compared to a typical analysis using a single database-search tool. KW - machine-learning KW - peptide-identification KW - peptide-identification-combiners KW - proteomics KW - tandem-mass-spectrometry AU - Searle, Brian AU - Turner, Mark AU - Nesvizhskii, Alexey PY - 2008/01/04/ UR - http://dx.doi.org/10.1021/pr070540w ER -