Max-Planck-Institut für Informatik
max planck institut
mpii logo Minerva of the Max Planck Society

Current Research Topics

Funded Projects / Cooperation

Computational Epigenetics

The demand for computational support and bioinformatic tools in the field of medical epigenetics is rapidly increasing, due to complex experimental methods, increasingly genome-wide analysis and the pressure to quickly translate scientific results into clinical practice. Our goal is to develop bioinformatic methodology for addressing these issues, and to implement a set of web services which make powerful algorithms available to typical bench scientists.

Protein Structure Prediction

In this area we are developing methods for predicting the structure of proteins given their amino-acid sequence. Our focus is on protein threading. With this technique the structure of the protein under investigation (the target) is modeled after a known structure of a suitable different protein (the template). We focus on the difficult cases, in which target and template share little sequence similarity (under 30%). In those cases the emphasis is on getting the backbone of the protein right. This project provides the protein structure prediction server ARBY that can be queried via the internet.
more ....

Relationships in Proteins

Protein structures are interesting mostly because they are the basis of the molecular function of the protein. Yet, to predict the protein's function from its structure is difficult and bioinformatics support of this task is less well developed than of protein structure prediction. In a first step, we provide a database STRuster a classification of differing structures of the same protein, occurring in the PDB, which we currently extend to identifying regions of structural difference and performing functional annotations of the respective structures.

Analysis of mRNA Expression Data

We are developing methods for analyzing mRNA expression data in order to gain insight into the functions of the involved genes. Our ScorePAGE algorithm maps differentially expressed genes to known biochemical pathways and, thus allows for scoring the relevance of pathways to reflect the differential expression of the involved genes. topGO (topology-based Gene Ontology scoring) is a software package for calculating the significance of biological terms from gene expression data. It implements various standard and advanced new algorithms for determining the relevance of Gene Ontology groups from microarrays.
more ....

Analysis of arrayCGH Data

During tumor progression, chromosome regions are deleted or amplified, resulting in genetic imbalances which promote uncontrolled cell proliferation. The arrayCGH data provides information on gene copy number changes, but the signal is often altered by experimental noise. We develop statistical algorithms for analysis of arrayCGH data in order to detect biologically and clinically relevant aberrations.
more ....

Medical Bioinformatics

Our work covers three overlapping research areas with projects concentrating on the analysis of molecular interaction networks and the prediction of protein function and interaction. We also develop and apply various bioinformatics methods to extract knowledge about the function and structure of proteins with medical relevance. Biological and medical cooperation partners identify promising candidate proteins that cause specific diseases, but are as yet relatively uncharacterized. The results of our bioinformatics analyses support the prioritization of experiments targeted towards elucidating the molecular function of the disease proteins. Currently, we focus on autoinflammatory and neurodegenerative disorders and viral infections. Examples are Crohn's and Parkinson's diseases and the hepatitis C virus.
more ....

Viral Recombination

We are developing methods for identifying individual recombination events in sets of molecular sequences. Our application background is recombination of viral pathogens such as HIV.
more ....

Docking and Drug Screening

We are developing methods and applying for docking small organic ligands - notably drugs - into active sites of proteins. Based on our successful work over about a decade we are extending docking methods to better score the docking solutions, better handle flexible protein pockets and increase the power of accuracy of docking into protein binding sites modeled by homology.
more ....

Analyzing metabolic networks in yeast

Here we are applying statistical and bioinformatics techniques in order to analyze metabolite data on yeast knockouts that have been fed with 13C-glucose. The goal is to better understand the molecular networks involved in amino-acid synthesis in yeast.
more ....

Bioinformatics for HIV

We are developing bioinformatic techniques for analyzing viral resistance of HIV to combination drug therapy.
The geno 2 pheno [resistance] server makes this analysis available via the internet. A new class of drugs, the coreceptor-antagonists, prevents HIV from entering and infecting immune cells. To assist in their administration, geno 2 pheno [coreceptor] predicts if a viral strain is likely to use a specific coreceptor. We are in the process of factoring host-guest interactions and structural aspects into our analyses.

Computational Chemistry

We are extending our docking technology to application outside biochemistry, notably to the problem of docking substrates into artificial receptors and we are optimizing such receptors