Profiling the metabolome using Fourier transform ion cyclotron resonance mass spectrometry, optimised signal processing, noise filtering and constraints methods

The main aim of this research study was to develop methods for metabolic profiling, a process involving the identification of metabolites present in biological samples, by applying Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS). Three novel methods were developed to improve the quality of mass spectra and their interpretation. The first, SIM-stitching, combined multiple, narrow spectra into a single, wide spectrum; an approach newly adopted by several laboratories in the MS and metabolomics communities. The second method employed a three-stage noise filter. Finally, the mass spectra were analysed using constraints optimisation methods and utility theory. It was discovered that SIM-stitching increased the effective sensitivity of the instrument five-fold, allowing many more metabolites to be detected. The three-stage filter approach showed how filter parameters can be selected to optimise noise reduction, and yielded significant benefits over other methods typically used. It was found that constraints methods can robustly and systematically identify the molecular formulae of compounds in a spectrum. A principal conclusion of this thesis is that optimised signal processing is essential to fully capture the metabolic content of samples. A further conclusion is that constraints methods can successfully be applied to metabolic profiling. Such methods have the potential to radically improve the quality of the metabolic results obtained from FT-ICR MS.

Name dc:type.qualificationname

d_ph

Level dc:type.qualificationlevel

d_ph

Grantor dc:publisher.institution

University of Birmingham

Year dc:date.issued

2011

Author dc:creator

• Payne, Tristan Graeme

• QC Physics