Data processing for metabolomic and proteomic profile data

 

Matej Oresic

 

VTT Biotechnology

Tietotie 2

P.O. Box 1500

Espoo, FIN-02044

Finland

Email: matej.oresic@vtt.fi

Phone: +358-9-456-4491

 

High-throughput integrated quantitative analysis of biological systems is becoming a desired approach to the study of functional interactions between genes, proteins, and metabolites. Recent such applications include studies of metabolic networks in S. Cerevisiae, energy transduction in Halobacterium NRC-1, and as a first such applications to the mammalian system, study of Atherosclerosis using ApoE3-Leiden mice as a disease model.

 

Analysis and comparison of molecular profile data across multiple samples has been an active field of recent research, and several statistical approaches to normalisation were developed enabling quantitative comparison of expression levels across different samples and genes. However, applications of techniques such as mass spectrometry and NMR to measure protein and metabolite levels in complex samples offer new challenges for data processing, integration, and interpretation. In order to obtain protein and metabolite levels several processing steps need to be applied, such as spectral alignment, peak detection, and normalisation.

 

Unlike in microarrays where gene identities are largely known, mass spectra require further steps for identification, such as MS/MS sequencing of tryptic peptides and subsequent peptide database searches for protein identification. Due to current instrumental limitations the identification process is very time consuming and only a fraction of spectral peaks can be associated with particular peptides or proteins, therefore a variety of computational methods need to be applied to select the peaks with highest potential relevants to the biological phenomena studied. As an example, we will present a recent study of ApoE3-Leiden mouse model of Atherosclerosis. We obtained simultaneous snapshots of gene expression, protein, and metabolic levels in two groups of mice, ApoE3-Leiden transgenic mice and an isogenic control group.