Identification of Peptides and Proteins in High Resolution Tandem Mass Spectrometry Data

Mass spectrometry has emerged as the leading technology for the identification and quantification of proteins in biological samples, playing an indispensable role in proteomic research. Even first steps towards clinical routine especially in terms of personalized medicine have been taken. Due to the complexity and the amount of generated data, specific software solutions are necessary to be able analyze them. Especially the identification of proteins and peptides from mass spectrometry data, one of the first but also one of the most important steps, is a challenging task. This doctoral thesis describes several algorithms for the analysis of such data sets, specifically designed to exploit the power of recent developments in instrument design, revealing high resolution and high accuracy data sets. A main part of this thesis is a new algorithm for database search, MS Amanda, capable of identifying peptides in mass spectrometry data. Applying MS Amanda leads to a higher number of identified peptides at the same false discovery rate compared to established software solutions. Additionally, algorithms have been designed for the identification of chimeric spectra -- spectra, carrying more than a single peptide --, revealing a potential, that otherwise remains unexploited. Even for data sets with instrument settings to avoid the occurrence of chimeric spectra, up to 30% of such spectra are measured, rising to 60% for complex samples. Up to 50% additional unique peptides, that would have remained unidentified, can be found at no extra measurement time applying a chimeric search. All results of this doctoral thesis have been disseminated through publication in internationally renowned journals and presentations at various conferences relevant for the proteomics community. All algorithms have been made available free of charge and are integrated in various software packages, enabling further downstream analyses of the identified spectra. These efforts had great impact on the international awareness of the algorithms presented in this thesis, also revealed by the number of citations and the number of downloads of the software.submitted by DI(FH) Viktoria Dorfer MSc.Universität Linz, Dissertation, 2019OeBB(VLID)356411