Improving Sensitivity in Shotgun Proteomics Using a Peptide-Centric Database with Reduced Complexity:  Protease Cleavage and SCX Elution Rules from Data Mining of MS/MS Spectra

Abstract

Correct identification of a peptide sequence from MS/MS data is still a challenging research problem, particularly in proteomic analyses of higher eukaryotes where protein databases are large. The scoring methods of search programs often generate cases where incorrect peptide sequences score higher than correct peptide sequences (referred to as distraction). Because smaller databases yield less distraction and better discrimination between correct and incorrect assignments, we developed a method for editing a peptide-centric database (PC-DB) to remove unlikely sequences and strategies for enabling search programs to utilize this peptide database. Rules for unlikely missed cleavage and nontryptic proteolysis products were identified by data mining 11 849 high-confidence peptide assignments. We also evaluated ion exchange chromatographic behavior as an editing criterion to generate subset databases. When used to search a well-annotated test data set of MS/MS spectra, we found no loss of critical information using PC-DBs, validating the methods for generating and searching against the databases. On the other hand, improved confidence in peptide assignments was achieved for tryptic peptides, measured by changes in DeltaCN and RSP. Decreased distraction was also achieved, consistent with the 3-9-fold decrease in database size. Data mining identified a major class of common nonspecific proteolytic products corresponding to leucine aminopeptidase (LAP) cleavages. Large improvements in identifying LAP products were achieved using the PC-DB approach when compared with conventional searches against protein databases. These results demonstrate that peptide properties can be used to reduce database size, yielding improved accuracy and information capture due to reduced distraction, but with little loss of information compared to conventional protein database searches.

A Simulated MS/MS Library for Spectrum-to-spectrum Searching in Large Scale Identification of Proteins

“…The simulated spectral libraries were based on the human IPI protein database v.3.29 (13) or a "decoy sequence database" where each protein sequence was read in reverse. A database of sequences was generated for tryptic peptides with mass between 900 and 4500 Da, allowing up to two missed cleavages and removing unlikely missed cleavage products (14). Charge forms up to MH 3 3ϩ were included for each sequence provided that a sufficient number of basic residues were present.…”

“…For this in silico generated peptide database, there are a total of 2,918,714 peptide sequences that were at least 9 amino acids long. Of these, 1,452,058 passed the missed cleavage rules of Yen et al (14). Simulation of spectra utilized the program MassAnalyzer (v.2.1), which generates the simulated spectra as DTA files.…”

“…Files were then formatted as MGF files for both X!Hunter and Mascot, and a converter was developed to change MGF-formatted experimental data to the MS2 format (15) used by BiblioSpec. X!Hunter XML and BiblioSpec SQT (15) output files were converted into an in-house MSPlus format (14).…”

“…Spectral library search applications require libraries that are preloaded into system memory. Because it is difficult to accommodate large libraries using 32-bit operating systems, searches against the simulated spectral library were performed by partitioning the input data set and library as described previously for peptidecentric database searching (14). MGF files of input experimental data and simulated spectral libraries were partitioned by charge and by MH ϩ with masses overlapping between adjacent partitions to accommodate the parent mass tolerance.…”

Crowdsourcing in proteomics: public resources lead to better experiments

“…First of all, cleavage may not always occur where expected (Hamady et al 2005;Thiede et al 2000;Yen et al 2006), and may even occur where it is not expected (Rodriguez et al 2008). Improved prediction of proteolytic behavior could therefore still benefit bottom-up proteomics approaches (Siepen et al 2007).…”