Improving sensitivity in proteome studies by analysis of false discovery rates for multiple search engines

Abstract

LC-MS experiments can generate large quantities of data, for which a variety of database search engines are available to make peptide and protein identifications. Decoy databases are becoming widely used to place statistical confidence in result sets, allowing the false discovery rate (FDR) to be estimated. Different search engines produce different identification sets so employing more than one search engine could result in an increased number of peptides (and proteins) being identified, if an appropriate mechanism for combining data can be defined. We have developed a search engine independent score, based on FDR, which allows peptide identifications from different search engines to be combined, called the FDR Score. The results demonstrate that the observed FDR is significantly different when analysing the set of identifications made by all three search engines, by each pair of search engines or by a single search engine. Our algorithm assigns identifications to groups according to the set of search engines that have made the identification, and re-assigns the score (combined FDR Score). The combined FDR Score can differentiate between correct and incorrect peptide identifications with high accuracy, allowing on average 35% more peptide identifications to be made at a fixed FDR than using a single search engine.

Augmented Annotation of the Schizosaccharomyces pombe Genome Reveals Additional Genes Required for Growth and Viability

“…Searches were performed using ProteinPilot (Shilov et al 2007) [version 3.0, settings: Search-Rapid; Instrument-OrbiFT MS (1-3 ppm), LTQ MS/MS; Enzyme-Trypsin; FDR Analysis-Yes] against the target database concatenated to its decoy database counterpart (Tang et al 2008). False discovery rates (FDRs) were calculated as described (Kall et al 2008;Jones et al 2009;Bitton et al 2010) (Table S7). Only peptides with confidence levels .95% were considered.…”

In-depth analysis of protein inference algorithms using multiple search engines and well-defined metrics

“…We analyzed the number of FDR filtered PSMs for each single search engine and their combinations before performing any protein inference evaluation. The benefit of combining search engine results for spectrum identification has already been shown extensively in other publications (44,45). It is generally accepted that search engines in combination yield more valid PSMs, especially in low-resolution fragment ion measurements (see section 1 in the Supplemental Files S3-S11).…”

Integrated Transcriptomic-Proteomic Analysis Using a Proteogenomic Workflow Refines Rat Genome Annotation

“…1). It is reported that peptides identified by two or more pipelines have fewer errors than the ones identified by only one pipeline (34,36). In fact, when we consider only the high confidence novel peptides (identified by two or more pipelines), performances of ITP, PA, and Enosi are similar.…”

“…1). The combination of multiple search engines and result integration increases both sensitivity and specificity in peptide discovery (33,34) and should be a beneficial practice in proteogenomics. It is developed upon the core of our previous automated prokaryotic analysis framework, GenoSuite, where four peptide identification algorithms, OMSSA (30), X!Tandem (31), InsPecT (35), and MassWiz (36) were configured for peptide identification.…”