Supporting data for "EXPANDING PROTEOME- COVERAGE : FROM MULTIDIMENSIONAL LIQUID CHROMATOGRAPHY TO BIOINFORMATIC"

<p dir="ltr">This thesis aims to enhance the in-silico identification and experimental characterization of rare PTMs by employing innovative bioinformatics methodologies and robust multidimensional liquid chromatography separation techniques. The first section presents an adaptive and fully automated bioinformatics tool for the validation and localization of rarely occurring PTMs. Utilizing a semi-supervised approach with a linear discriminant analysis (LDA) algorithm, it enhances verification through similarity scoring of tandem mass spectrometry (MS/MS) comparisons between modified peptides and their unmodified analogs. This tool addresses the limitations of traditional false discovery rate (FDR) control methods by incorporating orthogonal criteria, thereby improving the accuracy of PTM identifications. The methodology is validated through its application to a Macaca fascicularis model of stroke, resulting in the identification and confident validation of the largest number of endogenously nitrated peptides reported to date (Chapter 3). Furthermore, i its extensibility is demonstrated by its successful application to retrieve unidentified spectra, particularly for non-tryptic peptides commonly overlooked in traditional protein database searches, which are reported to account for 75% of yet-to-be-identified spectra (Chapter 4). The second section focuses on the development and evaluation of a streamlined multidimensional liquid chromatography (MDLC) system. The existing four-dimensional RP-SA(C)X-RP system, originally requiring four switching valves, has been reengineered. The new four-dimensional platform uses just two ten-port switching valves (2V); it maintains the original chromatographic resolving power and performance while reducing operational complexity and preserving solvent compatibility across separation dimensions [Anal. Chem. 2015, 87, 10015], leading to efficient peptide separation and increased detection of acidic, hydrophilic, hydrophobic and ion-suppressed peptides (Chapter 5). Key improvements include streamlined operational steps, minimized idle time, and simplified synchronization among four individual columns with complementary separation chemistries without the need to collect large numbers of fractions offline. The efficacy of the 2V-4D system was benchmarked through the analysis of the total lysate of the S. cerevisiae model, demonstrating its analytical performance in proteomic applications.</p>