Welcome to the bioinformatics and structural proteomics page!

In our Bioinformatics / Structural Proteomics group we investigate protein-protein interactions and protein structures.

Protein-structure and -function prediction is one of the most important fields in the post-genomic era. This is traditionally accomplished through the analysis and comparison of sequence and structure data. We have developed a novel non-homology method called "EMBed" that predicts structure and function based on interaction data. Furthermore, we have developed an algorithmic strategy that optimises the information gain on experimental effort in proteome-wide studies. Protein function can be described via interactions with other proteins or DNA (cellular function) and small molecules (biochemical function), hence the vague term of "function" can be understood mathematically in terms of Interaction-Graphs. In the context of the analysis of interaction data we are interested in the verification of experimental interaction data using bioinformatics methods, the mapping to functional units (domains) and the identification of signalling pathways. Based on the view that "Form-Follows-Function" the goal of our group is to explore protein interactions using three-dimensional structural information. A novel way of studying protein structure is to view proteins as graphs, where nodes are residues and edges contacts between them. We aim to find appropriate models to describe structures in this way. Such methods would have the additional advantage of allowing for conformational flexibility. Here we hope to apply the insights into biological networks we have obtained in the study of high-level interaction networks. Our main hypothesis (summarized very briefly) is: "Life is a Graph!"