Protein identification has been more helpful than before in the diagnosis and treatment of many diseases, such as cancer, heart disease and HIV. Tandem mass spectrometry is a powerful tool for protein identification. In a typical experiment, proteins are broken into small amino acid oligomers called peptides. By determining the amino acid sequence of several peptides of a protein, its whole amino acid sequence can be inferred. Therefore, peptide identification is the first step and a central issue for protein identification. Tandem mass spectrometers can produce a large number of tandem mass spectra which are used for peptide identification. Two issues should be addressed to improve the performance of current peptide identification algorithms. Firstly, nearly all spectra are noise-contaminated. As a result, the accuracy of peptide identification algorithms may suffer from the noise in spectra. Secondly, the majority of spectra are not identifiable because they are of too poor quality. Therefore, much time is wasted attempting to identify these unidentifiable spectra.<p> The goal of this research is to design spectrum pre-processing algorithms to both speedup and improve the reliability of peptide identification from tandem mass spectra. Firstly, as a tandem mass spectrum is a one dimensional signal consisting of dozens to hundreds of peaks, and majority of peaks are noisy peaks, a spectrum denoising algorithm is proposed to remove most noisy peaks of spectra. Experimental results show that our denoising algorithm can remove about 69% of peaks which are potential noisy peaks among a spectrum. At the same time, the number of spectra that can be identified by Mascot algorithm increases by 31% and 14% for two tandem mass spectrum datasets. Next, a two-stage recursive feature elimination based on support vector machines (SVM-RFE) and a sparse logistic regression method are proposed to select the most relevant features to describe the quality of tandem mass spectra. Our methods can effectively select the most relevant features in terms of performance of classifiers trained with the different number of features. Thirdly, both supervised and unsupervised machine learning methods are used for the quality assessment of tandem mass spectra. A supervised classifier, (a support vector machine) can be trained to remove more than 90% of poor quality spectra without removing more than 10% of high quality spectra. Clustering methods such as model-based clustering are also used for quality assessment to cancel the need for a labeled training dataset and show promising results.

Protein identification has been more helpful than before in the diagnosis and treatment of many diseases, such as cancer, heart disease and HIV. Tandem mass spectrometry is a powerful tool for protein identification. In a typical experiment, proteins are broken into small amino acid oligomers called peptides. By determining the amino acid sequence of several peptides of a protein, its whole amino acid sequence can be inferred. Therefore, peptide identification is the first step and a central issue for protein identification. Tandem mass spectrometers can produce a large number of tandem mass spectra which are used for peptide identification. Two issues should be addressed to improve the performance of current peptide identification algorithms. Firstly, nearly all spectra are noise-contaminated. As a result, the accuracy of peptide identification algorithms may suffer from the noise in spectra. Secondly, the majority of spectra are not identifiable because they are of too poor quality. Therefore, much time is wasted attempting to identify these unidentifiable spectra.<p> The goal of this research is to design spectrum pre-processing algorithms to both speedup and improve the reliability of peptide identification from tandem mass spectra. Firstly, as a tandem mass spectrum is a one dimensional signal consisting of dozens to hundreds of peaks, and majority of peaks are noisy peaks, a spectrum denoising algorithm is proposed to remove most noisy peaks of spectra. Experimental results show that our denoising algorithm can remove about 69% of peaks which are potential noisy peaks among a spectrum. At the same time, the number of spectra that can be identified by Mascot algorithm increases by 31% and 14% for two tandem mass spectrum datasets. Next, a two-stage recursive feature elimination based on support vector machines (SVM-RFE) and a sparse logistic regression method are proposed to select the most relevant features to describe the quality of tandem mass spectra. Our methods can effectively select the most relevant features in terms of performance of classifiers trained with the different number of features. Thirdly, both supervised and unsupervised machine learning methods are used for the quality assessment of tandem mass spectra. A supervised classifier, (a support vector machine) can be trained to remove more than 90% of poor quality spectra without removing more than 10% of high quality spectra. Clustering methods such as model-based clustering are also used for quality assessment to cancel the need for a labeled training dataset and show promising results.

<p> The mass spectra of disubstituted 1,4-dithiane, 1,4-oxathiane, and 1,4-oxaselenane, as well as the mass spectra of some disubstituted bicyclic and tricyclic sulfur and selenium compounds, were studied. The exact composition of the major ions of many of these compounds was determined. Fragmentation mechanisms are proposed to account for the major peaks in their spectra.</p> / Thesis / Master of Science (MSc)

<p> Reaction of the hydrated trichlorides of rhodium and Iridium with hexamethyldewarbenzene gave the complexes, (C5Me5MCl2)2 (M=Rh, Ir). A mechanism for this reaction is proposed.</p> <p> The dichloro complexes, (C5Me5MCl2)2, were reacted with a number of di- and tri-enes in ethanol in the presence of base and gave a variety of pentamethylcyclopentadienyl complexes of M(l) and M(lll). Evidence for a hydrido intermediate is presented and the hydrido- and deuterido- complexes, C5Me5IrH(D)ClPPh3, were isolated and characterised.</p> <p> The isomers endo-H and exo-H pentamethylcyclopentadiene(cyclopentadienyl) rhodium were isolated and showed significant differences in their properties. Cyclooctadienes reacted with (C5Me5MCl2)2 to give C5Me5M(1,5-C8H12) via the intermediacy of the π-2-cyclooctenyl complexes C5Me5MCl(C8H13); mechanisms are presented to account for the observed products. Cyclohexadienes gave the complexes, C5Me5M(1,3-C6H8). 1,4-Cyclohexadiene was isomerised to 1,3-C6H8; the rhodium complex, C5Me5Rh(1,3-C6H8), was a very active catalyst for the disproportionation of 1,3-C6H8 to cyclohexene and benzene, both ethanol and base were cocatalysts.</p> <p> Mass spectral data for these and other organo-metallic complexes are presented.</p> / Thesis / Doctor of Philosophy (PhD)

No description available.

No description available.

No description available.

Le typage bactérien est une mesure de contrôle essentielle pour lutter contre la diffusion des bactéries multirésistantes, mais les techniques actuelles sont longues et coûteuses. L'objectif de cette thèse était d'évaluer la capacité de la technique MALDI-TOF MS à typer les 3 principales espèces bactériennes responsables d'infections nosocomiales - Escherichia coli, Staphylococcus aureus et Pseudomonas aeruginosa. L'analyse par MALDI-TOF MS permet d'identifier les souches de E. coli appartenant au phylogroupe B2, souches les plus virulentes parmi les souches extra-intestinales, et au STI 31, clone très impliqué dans la dissémination des P-lactamases à spectre étendu. Chez S. aureus, cette technique reconnait les souches appartenant au CC398, impliquées dans une proportion importante d'infections graves chez des personnes fragiles. La technique MALDI-TOF MS identifie également cinq clones de P. aeruginosa à haut risque épidémique - STI 11, STI 75, ST235, ST253, ST395. Si nous avons confirmé des pics caractéristiques du phylogroupe B2 décrit également par d'autres auteurs, les pics biomarqueurs identifiant E. coli ST131 ou des clones épidémiques de S. aureus varient suivant les études. Différents paramètres peuvent influencer les résultats de typage par MALDI-TOF MS et doivent donc être standardisés. La technique MALDI-TOF MS permet d'identifier certains clones épidémiques. En gardant à l'esprit que le choix d'une méthode de typage doit être fait en fonction des objectifs mais aussi des performances des différents systèmes disponibles, la technique MALDI-TOF MS pourrait se positionner comme un outil de typage de première ligne. / Bacterial typing is crucial to tackle the spread of bacterial pathogens but current methods are time-consuming and costly. The objective of this study was to evaluate the ability of MALDI-TOF MS to type the 3 main bacterial species re.sponsible for nosocomial infections - Escherichia coti, Staphylococcus aureus and Pseudomonas aeruginosa. Analysis of MALDI-TOF mass spectra allow the identification (i) of E. coti isolates belonging to phylogroup B2, that fosters the most virulent extra-intestinal strains, and (ii) of E. coli STl 31, a clone involved in the dissemination of extended-spectrum β-lactamases. MALDI-TOF MS also recognizes S. aureus isolates belonging to CC398, a pathogen responsible for invasive infections in fragile patients and associated with a higher 30-day mortality. Furthermore the MALDI-TOF MS based typing method identifies the major high risk epidemic clones of P. aeruginosa STl 11, STl 75, ST235, ST253, and ST395. We confirmed phylogroup B2 specific peaks also described by other authors. However the identification of E. coli STl 31 or epidemic clones of S. aureus is based on biomarkers that differs between studies. Different parameters can influence the MALDI-TOF MS typing results and must therefore be standardized. MALDI-TOF-based typing methods allow the detection of epidemic pathogens. Keeping in mind that the choice of a method for routine bacterial typing should be made according to the objectives but also the performance of the different systems available, the MALDI-TOF MS technique could become a first-line typing method.

A new experimental approach is reported that focuses on photoionization mass spectrometry of primary photofragments. Photodissociation of molecules and radicals is carried out by using a tunable, pulsed dye laser. Nascent photofragments are subsequently photoionized by time-correlated vacuum-ultraviolet (VUV) radiation, so that only single-photon ionization occurs. Several experimental approaches are used in order to optimize pulsed radiation sources in the VUV that are suitable for pump-probe experiments. Laser systems that rely on frequency tripling in suitable non-linear media are used as an intense VUV radiation source of narrow bandwidth and limited tuning range. Monochromatized laser-produced plasma radiation is also used, which provides tunable VUV radiation throughout the entire VUV energy regime (8-25 eV). Ionized photofragments are finally detected by time-of-flight mass spectrometry. This experimental approach is used in order to investigate the following issues: * Absolute photoionization cross sections of atomic and molecular photofragments are derived from photoionization mass spectrometry. This is shown for the photolysis of chlorine dioxide (OClO), chlorine monoxide (ClO) and ozone (O3). The latter species generates O(1D), so that photoionization and autoionization of this excited species was studied for the first time. * Quantum states and quantum state distributions are derived from photoion yields of primary photofragments in order to characterize photodissociation processes. This is shown for the predissociation of OClO, which yields vibrationally excited ClO in its electronic ground state. * Branching ratios and quantum yields of competing photochemical pathways are determined from photoionization mass spectrometry. The accuracy of this approach is superior to other techniques. These investigations allowed us to investigate photolysis processes of molecules and radicals that are of significant importance with respect to recent issues of atmospherical photoprocesses, such as polar ozone depletion.

Halogenhaltigen atmosphärischen Spurengasen wird eine Schlüsselrolle bei lokalen, regionalen und globalen Veränderungen der Erdatmosphäre zugesprochen. Die Photolyse dieser Stoffe durch ultraviolette Strahlung der Sonne führt zum Eintrag reaktiver Atome und Radikale in die Atmosphäre. Dies betrifft vor allem den stratosphärischen Ozonabbau, der nach heutigem Kenntnisstand durch Reaktionszyklen katalysiert wird und bei dem die Photolyse halogenhaltiger Spurengase an zentraler Stelle beteiligt ist. Im Rahmen dieser Arbeit wurden primäre Photoprozesse halogenhaltiger atmosphärischer Spurengase in Laborexperimenten charakterisiert. Dabei bestand das Ziel in der Bestimmung von primären Quantenausbeuten und Verzweigungsverhältnissen konkurrierender Photolysekanäle. Hierfür wurde ein neuartiger experimenteller Ansatz genutzt, der die folgenden Komponenten beinhaltet: (a) Photolyse der Spurengase durch gepulste ultraviolette Laserstrahlung unter stoßfreien Bedingungen, (b) Ein-Photon-Ionisation der neutralen Photolyseprodukte mittels durchstimmbarer Vakuum-UV-Strahlung und (c) Identifizierung der gebildeten Photoprodukt-Ionen durch Flugzeit-Massenspektrometrie. Dieser Ansatz ermöglicht die Identifizierung aller gebildeten Photolyseprodukte, die Bestimmung des Anregungszustandes der Photoprodukte durch Ausnutzung von Autoionisationsprozessen und die Bestimmung von Verzweigungsverhältnissen und Quantenausbeuten konkurrierender Photoprozesse.Im einzelnen wurden die atmosphärischen Spurengase Chlormonoxid (ClO) und sein Dimer (Cl2O2), Nitrylchlorid (ClNO2), Brommonoxid (BrO) und Bromnitrat (BrONO2) untersucht. Die Resultate können zur Verfeinerung atmosphärischer Modelle und deren Prognosefähigkeit beitragen. Insbesondere können die Ergebnisse auch einen Beitrag zur zuverlässigen Interpretation von Ergebnissen aus Feldstudien leisten.