Methods for mass spectrometric proteome analysis

The major goal of proteome analysis is structure determination, identification, estimation of expression level, and understanding of the role of any protein in an organism. In combination with genomics, proteomics can provide a holistic understanding of the biological processes occurring in any organism. Mass spectrometry-based proteome analysis typically utilizes mass spectra of peptides of digested proteins together with sequence collection searching for rapid and accurate identification of proteins. Successful proteome analysis requires good experimental design, high quality data and optimized search conditions for protein identification. A mass spectrometry-based method for differential detection and identification of proteins in protein mixtures utilizing multivariate methods was developed. The method utilizes intensity values from matrix assisted laser desorption/ionization time-of-flight mass spectra of tryptically digested protein mixtures for the label-free identification of a protein present in different concentrations in two samples. The Probity algorithm, which assigns the statistical significance to each identification result, was applied for the protein identification. A systematic study of the quality of peptide mass fingerprint based (PMF) protein identifications under different search constraints was performed. 2244 PMFs from 2-dimensional gel electrophoreses separated human blood plasma proteins were submitted to the Probity algorithm for protein identification under different search conditions. The number of significantly identified proteins was counted for each condition in order to find the best set of search constraints for successful outcome. A study of how the quality of proteolytic peptide identification can be improved by optimizing the information content of tandem mass spectra and by optimizing the search constraints of the sequence collection searching was done. The X! Tandem algorithm was employed for identification of proteolytic peptides from mouse proteins. The influence of the mass accuracy of both precursor and fragment mass ions, the number of sequences included in the search, and the number of missed proteolytic cleavage sites on the number of identified peptides was explored. Computer simulations were performed in order to investigate quantitatively the information content in tandem mass spectra of proteolytic peptides, required to identify peptides and their post-translational modification.