Novel algorithms adding new dimensions to mass-spectrometry based proteomics: comprehensive characterization of posttranslational modifications

The technological advances in proteomics are allowing an increasingly detailed characterization of the complex panorama of post-translational modifications of proteins and are gradually developing towards an unbiased analysis of peptide modifications. The recently developed ultra-tolerant database search (open search, “OS”) uses precursor mass tolerances of hundreds of Daltons, allowing the identification of modified peptides never identified before by conventional (closed, “CS”) searches. Despite these improvements, OS algorithms still rely on the chance that the modification leaves enough unaffected fragment ions, thus identifies only half of the modified peptides and cannot pinpoint the modification site. Furthermore, there is a need of a generic quantification algorithm able to handle the huge variety of modified peptides resulting from an OS experiment. In this Thesis, I present a suit of developed algorithms and tools, designed to overcome the above-mentioned limitations. Comet-PTM is an improved search engine that applies the peptide modification mass to the fragmentation series upon score calculations for each peptidespectrum match (PSM). As a result, we emulate the scores produced by a CS for the same modification set as variable; double the yield attained by a regular OS and localize the modified residue with high accuracy. SHIFTS, controls the PSM false-discovery rate of the CometPTM results through a conservative three-layered approach taking into account the high mass accuracy of modern mass spectrometers. PtmSticker annotates the enormous wealth of modifications in a semi-supervised way, allowing for the first time the generation of a complete map of the modified peptidome as part of an automated pipeline. For the quantitative analysis of modified peptides, we developed and validated an algorithm based on a previously proposed WSPP workflow, for the simultaneous quantification of the modified peptidome, the whole proteome and systems biology. The model allows detection of PTMs changing independently of the protein abundance change. These developments were used to characterize the impact of mitochondrial heteroplasmy on the proteome and on the modified peptidome in mice, revealing that the heteroplasmy causes oxidative damage in heart OXPHOS proteins.