Computational Methods in Medical Genetics and Expression Data Analysis

Part of a gene disruption network based on gene expression data, figure by Kimmo Palin


See also

Collaborators outside the Unit

The group develops methods and tools for the analysis of different types of genetic data (marker and expression data). The results include novel techniques for inferring possible genetic models from population risk parameters; a related result shows that for any multilocus model of autosomal inheritance the sibling risk is always at least as large as the offspring risk. The group has also developed new algorithmic methods for haplotype association analysis and for finding correlations between spatial and expression profiles in genomes.

Another tool finds correlations between the expression profiles and the patterns of symbols in the regulatory areas of genes. This tool which is capable of processing entire genomes can be used for finding binding sites of the so called transcription factors.

The developed tools include also an extensive simulation methodology for analyzing significance of findings in genetic studies and for study design in isolated populations.

Plans for the next years include extensive tools for the study of inheritance of gene expression and for the study of regulatory effects of spatial location. A particularly challenging task is the integrated analysis of genome, proteome and metabolic expression data. The ultimate goal is the synthesis of gene regulatory and metabolic networks that model the functioning of a cell. Biologically realistic and efficient algorithms are much in need in all these applications. The development of the tools and methods will take place in close cooperation with our collaborators in application areas.