We center around Bioinformatics and Systems Biology. Our research covers the main branches signal processing, pattern recognition (data mining: clustering and classification), and visualization. Signal processing includes especially the analysis of time series data and the processing of single images and (volumetric) image sequences. Challenges in the biomedical field are e.g. clustering and classification of high-dimensional data of low sample size (HDLS).
Receiver operator characteristic (ROC) curves are recommended to assess the diagnostic value of tests depending on a single cutoff value of a continuous variable. These ROC curves show the true positive rate (sensitivity) against the false positive rate (1-specificity). It is desirable, especially in situations with small samples of observations, to display confidence bounds of the ROC curve. A Perl module was programmed which calculates the ROC curve and its distribution free confidence bounds. A simple user interface also written in Perl permits their display. The program is available at http://www.CPAN.org.
GExVis is a convenient tool for the interactive and intuitive exploration of gene expression data. The software implements a graphical representation of gene expression data in a new way. The special color coding scheme enables the user to simultaneously observe the absolute expression values in conjunction with its ratios to a free selectable cell line. In this way biologically relevant differences can be visually explored. The software is, due to its implementation as a Microsoft Excel add-in, easily accessible and usable. The software is freely available from the author upon request. A manuscript is submitted.
MCGH-Analyzer is a user friendly software (Microsoft Excel based) for array-CGH analysis. Array-CGH measures genomic aberrations. It was developed together with the working groups of Prof. Bentz (Städtisches Klinikum Karlsruhe), Prof. Döhner (University Hospital Ulm), and Prof. Lichter (German Cancer Research Center - DKFZ, Heidelberg). This Excel add-in has a Visual-Basic front-end and a C++ library to perform calculations, links to a Java based ideogram browser and R also exist. The software is freely available from the author upon request. A manuscript is in preparation.
Originally part of MCGH-Analyzer the IdeogramBrowser software is now available as a standalone program. It is a freely available Java application for visualizing genomic abnormalities. The program is available at IdeogramBrowser Software.
Getting an overview over complex dependencies among set relations (e.g. gene lists) is often a difficult task. Standard tree representations are in many cases an improper choice for this issue, especially in representing intersections. A Venn diagram representation of gene sets could reveal more valuable information to the researcher. Full containment of one set into the other, partial intersection and disjunctness can be seen with a glance with Venn diagrams. We propose a Venn representation using polygons with areas directly proportional to the true cardinalities of the set to show the maximum possible information in a 2-dimensional drawing. VennMaster is a small and easy to use, platform independent, Java application which represents set relations by polygons. It allows an interactive exploration of the sets. VennMaster was tested on Windows XP, Linux and Mac OS X using the java runtime environment 1.4.2 (http://java.sun.com). The program is available at VennMaster Software.