| Development of database search algorithms for protein identifications and how to improve the performance of the algorithms are important research areas in the mass spectrometry-based proteomics.; In this thesis, two widely used database search algorithms, SEQUEST and X!Tandem, were studied in detail. The strengths and weaknesses of SEQUEST and X!Tandem were studied, and peptides identified and missed by these two algorithms were characterized and compared in terms of the charges the peptides, the amino acid compositions of the peptides and the lengths of the peptides, etc. Research results show that SEQUEST is relatively prone to identify single charged peptides, while X!Tandem is prone to identify highly charged (triply charged, for example) peptides. Both SEQUEST and X!Tandem identify peptides reasonably well for peptides that show selective cleavages and neither of them seems to be good at identifying short peptides with few amino acid residues. (Abstract shortened by UMI.)... |
