Improving 2DE proteomics by computerized image analysis

Summary

The science of functional genomics comprises investigations of the pathways spanning from the genetic composition to the final physiology of an organism. For all living organisms on the earth, the genes carry the information necessary for reproduction and regulation. Each individual has its own genetic makeup, giving rise to differences between individuals. The atom of a biological system is the cell. Its complex machinery reads genetic information (genomics) that is transcripted through RNAs (transcriptomics) to synthesize proteins (proteomics) and metabolites from enzymatic activities (metabolomics). The “-omics” aims at the collective characterization and quantification of pools of biological molecules that translate into the structure, function, and dynamics of an organism or organisms. Whereas the sequences of the genetic code are more or less constant throughout the whole life period of an organism, the genes are constantly turned on and off in a dynamic process as a response to external and internal environmental factors as well as to the developmental stage of the organism. When a gene is turned on, a copy of the gene is produced containing the code for a specific protein. The proteins have different functions in the cells such as being enzymatic proteins executing chemical reactions, hormones that give signals to other cells, or structural proteins used as building blocks for cells or organs among others. Thus, the activation of genes and the following protein synthesis will in turn determine the metabolic activity in the cells, and thereby the final physiology of the organism. These -omics, in addition to many other -omics such as interactomics, lipidomics, glycomics, immunomics, epigenomics, phenomics, etc. can be seen as dynamic multidimensional biology to apprehend the cell and/or organism as a whole integrated, heterogeneous, interacting and dynamic system. Proteomics sits within this -omics world and enjoys the hype formerly generated by genomics to supplement, along with other -omics, information about the dynamic state of the biological activity that genomics lack to provide. One tool for analyzing proteome is two-dimensional electrophoresis (2DE) where prteins are separated in a gel in two dimensions by different physical properties (most often size and pH), which results in images of spots. Traditionally, each individual 2DE image, representing one replicate of one biological sample, is first segmented into its many different spots and the volume is quantified for each spot in each gel image; thereafter lists of protein spot volumes from different samples are collected for statistical analysis. This segmentation-before-analysis approach is known to cause considerable segmentation problems due to weak or overlapping protein spots, which in turn causes, for instance, missing values and other misrepresentations in the resulting spot volume table and hence problems in the statistical analysis.