Preprocessing strategies for DDA/DIA LC-MS/MS quantitative molecular profile data
This thesis explores algorithmic improvements for liquid-chromatography in tandem mass-spectrometry (LC-MS/MS) data quantification and its applications on a proteomics context. The main contribution of this work is the PASTAQ pipeline, which aims to quantify and pre-process LC-MS/MS data without the use of arbitrary intensity thresholds, enabling the detection of lower concentration compounds in the same data compared with other state-of-the-art tools.
We explored how some algorithms could be generalized to other similar technologies, like matrix-assisted laser desorption/ionization (MALDI), and how some of the algorithms could be generalized and sped-up by the use of graphic process units for massively paralell computing. Additionally, we used the pipeline on a practical application for the study of histone acetylation with a combined approach of metabolic and chemical labeling (CoMetChem).