| Popis: |
The development of analytical technology over the last few decades has opened the door to new and intriguing fields of research, allowing investigations at a level of quality not achieved before. Chromatography and mass spectrometry have undergone a revolution of advances in the last four decades with extensive developments in new stationary phases and new developments in ionisation and ion separation techniques. The work in this thesis investigated the application of two of these advances, hydrophilic interaction liquid chromatography (HILIC) and Fourier transformation mass spectrometry (FTMS) for the analysis of complex biological samples. This thesis focuses on the development of metabolic profiling methods which have applications in disease diagnosis, personalized healthcare and drug discovery. The complex nature of the type of samples is discussed and data simulation was used to show the importance of a separation step before detection, even with high resolution mass spectromet ers. HILIC was compared to more traditional reversed phase stationary phases and found to be more suitable for the research questions in this thesis. Several commercially available chromatographic columns share the HILIC retention mechanism but differ in their chemistry. Four of these columns were compared using injections of 140 authentic standards and they were evaluated according to the retention range for the standards, peak width, isomer separation and background signal level generated in the mass spectrometer. The most "fit for purpose" column was then selected and further tested with different organic modifiers and mobile phase additives. The final method was tested for repeatability and reproducibility of retention time. The selected analytical method was then applied to the analysis of the metabolome of two biological systems i.e. Drosophila melanogaster and Trypanosoma brucei. This demonstrated the ability of this method to rapidly profile and extract biological information from the tested systems. The complex output of the analytical methods needed data processing tools, and for this purpose a computer programme was written and is described and tested in the work reported in this thesis. The in-house built code was compared with a commercially available and an academic software package and the pros and cons of each of these tools are discussed within this thesis. |
