Peptide sequence assignments by probabilistic peptide profile matching to an annotated peptide database

One goal of proteomics is to identify and catalog the proteins present in biological systems to aide in mapping protein-protein interactions and protein expression patterns. The global analysis of enzymatically digested cell extracts via liquid chromatography coupled with tandem mass spectrometry has become a powerful and popular method for accomplishing this goal. However, one challenge encountered when using LC-MS/MS is the identification of lower abundance proteins. This challenge has been previously addressed by repeated sample analysis which can be problematic when analyzing biological samples where the sample is precious and amounts are limited, or where instrumentation time and resources are limited. This document presents a method of peptide and protein identification from LC-MS analyses that generates peptide sequence assignments by matching the LC-MS derived chromatography and mass property coordinates to those of previously identified peptide sequences archived in an annotated peptide database. The matches are scored according to the probability of an overlap between the peptide feature and the database peptide coordinates. These scores are used to create a ranked list of possible peptide sequences for each peptide coordinate pair submitted. The method resulted in over six times more peptide sequence identifications from a single LC-MS analysis of yeast than from direct peptide sequencing using LC-MS/MS. To define the chromatography and align the coordinates for multiple runs, a method of chromatography normalization and gradient alignment was created by monitoring the signal intensity of mobile phase gradient tracers and run-to-database chromatography normalization using stepwise linear approximations. This gradient monitoring method chronicled the random distortions that can affect chromatography and that are often encountered and undetected during reverse phase LC-MS and LC-MS/MS analyses. The method was also found to better define peak positions than the traditionally used retention time and gradient composition. Normalizing the reverse phase chromatography to the annotated peptide database as reference decreased the run-to-run variability in chromatography values, thus aiding in the improved identification by peptide profiling using the chromatography and mass characteristics.; In addition, the document briefly describes some of the rationale and advantages of peptide profiling compared to tandem mass spectrometry, and some of the factors contributing to this difference.