Data Mining in Proteomics

In the past decade, major developments in instrumentation and methodology have been achieved in proteomics. For proteome investigations of complex biological samples derived from cell cultures, tissues, or whole organisms, several techniques are state of the art. Especially, many improvements have been undertaken to quantify differences in protein expression between samples from, e.g., treated vs. untreated cells and healthy vs. control patients. In this review, we give a brief insight into the main techniques, including gel-based protein separation techniques, and the growing field of mass spectrometry. The proteome describes the quantitative expression of genes within, e.g., a cell, a tissue, or body fluid at specific time points and under defined circumstances (1). In contrast to the genome, the proteome is highly dynamic and the protein expression pattern of cells in an organism varies depending on the physiological functions, differentiation status, and environmental factors. In addition, alternative splicing of mRNAs and a broad range of posttranslational modifications (e.g., phosphorylation, glycosylation, and ubiquitination) increase proteome complexity (2, 3). Transcription analysis also does not allow insight into degradation and transport phenomena, alternative splicing, or posttranslational modifications. Furthermore, mRNA and protein levels often do not correlate (4, 5). All these influences are unconsidered in genome analysis and underline the importance of proteome analysis to obtain deeper insights into cellular functions. In general, proteome analysis provides a snap-shot of proteins expressed in a cell or tissue at a defined time point (1). Indeed, not only qualitative analysis resulting in a defined “protein inventory”