Challenges in mass spectrometry: computational lipidomics and MS imaging (R-8547)

Mass spectrometry is a versatile analytical technique for identification of molecular compounds by measuring their mass to charge ratio. Intuitively, it is a method of "weighing" individual molecules. One of many research areas which use mass spectrometry is proteomics, the study of proteins. Mass spectrometry can be used, among others, to identify proteins in a biological sample; obtain the sequence of a protein; infer the presence and location of post translational modifications, or gain insight into the structure of a protein. That said, the mass-to-charge ratio of the whole protein does not give much information about it, and therefore, the proteins are usually fragmented before being analyzed. Even though masses of two different proteins can be very similar, the masses of their fragments are often different because of different amino acid compositions. Applications of mass spectrometry to proteomics and techniques of protein fragmentation are relatively well studied. Much less is known are applications about lipid samples, even though lipids constitute one of the most fundamental groups of biological compounds. The presence of double bonds and functional groups makes this group of biomolecules extremely diverse, with hundreds of thousands of different molecules. In this project, we will translate the vast expertise of both research groups on mass spectrometry-based proteomics into field of mass spectrometry-based metabolomics and lipidomics. The first part of this project will be developing computational tools to aid analysis of lipid samples using the mass spectrometer. In particular, we will focus on modeling fragmentation pathways to help understand how to perform optimal fragmentation of lipids. We will develop a software tool to infer the quantities of different Special Research Fund Application in the framework of the Bilateral Cooperation Programme April 2017 fragments in a spectrum, which will allow to estimate the fragmentation patterns and correlate them with instrumental settings. We will also investigate the newly emerged technique called Mass Spectrometry Imaging. This technique is based on obtaining mass spectra from many points of a tissue sample, and then plotting the intensities of products on an electron micrograph of the sample. This way, we can plot the concentrations of many compounds by performing a single experiment that overlay with high-resolution microscopy images that allow for a multimodal interpretation of the data. However, the huge amounts of data generated this way pose many challenges in employing the data analysis and interpretation of such hyper cubes. We will develop tools for processing, normalization, and data mining of mass spectrometry imaging data to allow for a more convenient use of this multi-modal mass spectrometry imaging.

Keywords:CLUSTERED DATA

Disciplines:Applied mathematics in specific fields, Statistics and numerical methods