Analysis of Specific Metabolitesmetabolomics_target

Our group is involved in the ongoing development and improvement of methods for the target analysis of metabolic biomarkers of oxidative stress and neuronal damage. Quantitative determinations are performed employing UPLC-ESI-QqQ and are carried out in urine, plasma and tissue extracts.

  • Protein oxidation: m-tyrosine/p-tyrosine, 3-Nitro-tyrosin/p-tyrosine, 3-Cl-tyrosine/p-tyrosine, o-tyrosine/phenylalanine, 3I-Tyrosine
  • DNA oxidation: 8-oxo-desoxyguanosine/2-Desoxyguanosine
  • Cholines: citicoline, acetylcholine and choline
  • Biomarkers of neuronal damage: eight isoprostane isomers
  • Oxidative stress biomarkers (transsulfuration pathway): glutathione (GSH), glutathione ox. (GSSG), cysteine, cystine, methionine, S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH), homocysteine, homocysteine  ox., cystathionine, g-glutamylcysteine, g-glutamylcysteine ox.

We are constantly improving our analytical determinations and the developemtn of methods for the determination of novel biomarkers is in due course.

Un-targeted Approachesmetabolomics_untarget

The over-all goal of metabolomics is the global assessment and validation of endogenous metabolites within a biologic system. We are developing UPLC-MS/MS methods dedicated to characterizing biological fluids such as urine or tissue extracts (e.g. retina). With the help of multivariate statistical tools, we try to identify new biomarkers that are clinically relevant. Human samples as well as samples from animal-experiments are under study.

Data Analysisdata_analysis

In untarget metabolomics studies, complex data sets with several hundreds or thousands of measured variables (e.g. concentrations of analytes, spectroscopic signals, etc.) are obtained for each sample. In such cases it is not straightforward to assess each variable separately. Basically, in bioanalytical analysis, two fundamentally different situations can be distinguished, requiring specific data analysis tools: (i) pattern recognition (classification) and (ii) multivariate calibration (quantification).

When applying multivariate statistical tools, special emphasis is put on the selection/elimination of variables as well as data pre-treatments (e.g. normalization, derivatives, etc.).

Currently, in our group we are applying multivariate data analysis tools for gaining a deeper insight into complex data and revealing novel biomarkers using Matlab 7.7.0 from Mathworks (Natick, MA, USA).