Search
Links and Functions
Language Selection
Breadcrumb Navigation
Main Navigation
Content
Structural Bioinformatics
Since it is the structure of a protein which finally accounts for the distinct functional properties like enzymatic reaction mechanisms and the specificity of protein-protein interactions those can only be understood in detail on the structure level. Novel developments in genome sequencing techniques and experimental biology have led to an exponential increase in the number of known protein sequences (for which no protein structures are available) but also to an strong (nevertheless much slower) increase of available protein structures in the last years. In order to assign many protein sequences with unknown structures to their potential structures based on involved sequence comparison techniques and also to analyze existing protein structures and their relationships in more detail, we have developed a large set of methods, databases and software tools.
Research Team
Structure-based analysis of alternative splicing
Alternative Splicing and Protein Structure Evolution
Supplementary Material
Alternative Splicing is an important process in eukaryotic cells contributing significantly to functional diversity of higher organisms as more than one gene product can be produced from one genomic locus. In this study we have analyzed alternative splicing events on the protein structure level and describe interesting implications of a joint analysis of splicing and protein structure data for the analysis of protein structure evolution and functional diversity generated by alternative splicing events. The results of this study have been presented as Highlights Track Paper at the ISMB 2008 in Toronto and the GCB 2008 in Dresden. The slides of the talk at the ISMB can be found
here
ProSAS - Analyzing alternative splicing in the context of protein structures
Database Webinterface
In order to support the large-scale and genome-wide analysis of alternative splicing on the protein structure level, we have build the ProSAS database which is a comprehensive database containing structure and splice annotations for various genomes (currently Human, Mouse, Rat and Chimpanzee) as well as all proteins with splicing annotation in Swissprot. Via the intuitive web interface users can analyze the expected structural effects of a splicing event on a protein's structure and compare splice variants originating from the same gene according to their structure and function.
Protein structure comparison and analysis
Vorolign
Webserver
Vorolign is a highly accurate and very fast method for pairwise and multiple protein structure alignment and automated protein structure classification. The method is available as a Webserver and as standalone version upon request. The Vorolign Paper has been awarded the Best student paper award at ECCB 2006 in Eilat.
PPM
Online Material
PPM is a novel protein structure alignment procedure explicitely modelling and taking into account structural flexibility often observed in members of the same family (phenotypic plasticity). The method is able to produce highly accurate alignments of conserved structural cores while trying to score the evolutionary distance according to the observed structural mutations.
Large-scale comparison of SCOP and CATH
Project Pages
With this approach compare SCOP and CATH according to different aspects and extract valuable knowledge from the orthogonal knowledge contained in both datasets
AutoPSI
Database
AutoPSI is a large-scale application of methods developed in the group for an automated prediction of SCOP families to thousands of unclassified protein sequences and structurs.
Protein structure prediction
SSEP-Domain
Webserver
SSEP-Domain is a powerful method for predicting domain boundaries and domains in protein sequences of unknown structure. A server version of the program can be found here .
AutoSCOP
Webserver
AutoSCOP is a accurate, pattern-based method to predict the SCOP classification of a protein sequence based on InterPro member database patterns being unique for a specific level of the SCOP classification.
Secondary Structure Prediction based on frequent patterns
Software available upon request
With this approach based on frequent patterns we have proposed a novel, potentially window-free feature representation for protein secondary structure prediction. The method is build around a modification of the Apriori algorithm to mine frequent patterns (potentially composed of regular expressions) in a database of sequences. The PERL implementation of the algorithm can be found here TODO .
QUASAR
Java Application
QUASAR is a flexible framework to optimize scoring functions for the assessment of sequence-structure alignments. The software is written in Java and available for download and via Java-Web-Start here .
Experimental protein structure determination
Analysis of chemical Shift data for NMR structure determination
Software and Services
The goal of this project is to achieve a considerable speed-up in the NMR structure solving process. Therefore computational methods are developed which allow the creation of models for a protein structure at an experimental stage in which this is not possible by hand. These models may then be refined and tested for correctness using the "standard" way. The time saving results from the need for fewer experiments in when building the initial model.
2007
Papers
Simon W. Ginzinger , Thomas Gräupl , Volker Heun .
SimShiftDB: Chemical-Shift-Based Homology Modeling .
Sepp Hochreiter , Roland Wagner (eds.):
Proceedings of the First International Conference on Bioinformatics Research and Development (BIRD'07), Berlin, Germany, March 12-14, Lecture Notes in Bioinformatics, vol 4414, pp. 357-370, Springer, 2007.
2006
Automatic correction of inconsistent chemical shift referencing
Software and Services
The construction of a consistent protein chemical shift database is an important step toward making more extensive use of this data in structural studies. Nowadays available data is frequently impurified, in particular with respect to chemical shift referencing, which is often either inaccurate or inconsistently annotated. CheckShift is at tool for preprocessing chemical shift data to detect and correct referencing errors.
Service Menu
Footer
