• Top
• Oral presentation

Oral presentation

Development of a new generation of drugs will significantly benefit from the huge amount of genomic and expression data that are generated these days if, and only if, state-of-the art methods of bioinformatics are applied for their interpretation. As a real interdisciplinary science, bioinformatics have to make proper use of the existing biological knowledge and to combine it with sophisticated algorithms to find (practicable) solutions of pharmacological problems. Such knowledge-based approaches can give momentum to the so-called "wet-dry cycle" of basic as well as applied research in modern life sciences.

One problem for genome-based pharmacological research is that we have to cross the borderlines between several levels of complexity, e. g. those between one-dimensional genomic information storage and the multidimensional molecular organization with a cell, or that between the the cellular and the physiological level of organization. Classically, bioinformatics focused on storing sequence data, assembling genomic fragments, and finding open reading frames within the genomic sequences. It took some time that revealing the signals that provide the dynamic properties of the genome, i. e. the expression of the genes, gained broader interest. It became now an important part of "functional genomics" approaches. Based on the knowledge stored in corresponding databases (e. g., TRANSFAC®, the vocabulary and the syntax of genomic sequence signals that govern gene expression and the molecules recognizing them is going to be understood. The molecules (transcription factors) are frequently end-points of signal transduction cascades and pathways. To represent these regulatory networks in a computer-manageable manner and to render them amenable to computer-aided simulations is another big challenge to bioinformatics. Together, all these transcription regulatory mechanisms represent the way how the one-dimensional genomic information is "translated" into cellular reality.

Experimental support for the bioinformatics attempts to characterize and identify transcription regulatory sequence signals comes from gene expression mass data which, without profound promoter analysis, can be used only in a phenomenological way, their interpretation being restricted to modeling the effects of the genes found to be affected (induced or repressed) under certain conditions, rather than the causes of the phenomena observed. The advent of these new technologies generates mass data which even exceed mere genome sequence data by orders of magnitudes. To make systematic and comprehensive use of these data, they have to be mapped on the one side to the genome (i. e., the "system" of the genes) on the one side, and to a systematic representation of the organs/tissues/cell types where the gene expression data have been obtained from. For these systematic representations, "ontologies" are going to be developed such as GO for gene functions and CYTOMER® for anatomical/morphological structures, their developmental stages and physiological roles. With the help of these ontologies and the systematic determination of gene expression patterns and profiles, transcend the borderline between the cellular and the physiological level of organization.

For practical use, a bioinformatics system is required which allows to deal with data about processes on the different levels, to zoom between the genomic and the physiological level and to work transparently on those levels that are relevant for a specific problem. A concept and an implementation for such a system ("PheGe") will be presented.

Have something to say? Post a comment on this article!