09.00: Registration

10.00: Welcome, Martin Pohle, Olaf Brenn

Session 1:  New Developments and Bioinformatics Research

10.15: Topic 1: Dr. Steffen Neumann, Leibniz Institute for Plant Biochemistry Halle, www.ipb-halle.de

Plumbing Metabolomics

Mass spectrometry is an important analytical technology in metabolomics (and proteomics) research. The first steps in the analysis pipeline for MS measurements are signal processing and peak picking tasks. We are developing and enhancing XCMS and related Bioconductor packages. After identification of relevant features, a putative identification is based on mass decomposition to calculate the empirical formula. Further characterisation uses Tandem-MS libraries and spectral searching in compound libraries. The Components of the pipeline are connected via web services and workflows.

30 Min. Presentation + Discussion

11.15: Topic 2: Prof. Ivo Grosse, Martin-Luther-Universität Halle, www.informatik.uni-halle.de/arbeitsgruppen/bioinformatik

Modeling Complex Biological Systems by Variable-Order Bayesian Networks

One of the challenges of modern bioinformatics and systems biology is the integrative analysis of high-throughput data from diverse areas of biology including genomics, transcriptomics, metabolomics, proteomics, structural biology, molecular phylogeny, population genetics, or taxonomy. Analyzing and modeling these data in a holistic manner requires modern approaches from machine learning and information theory. One of the most popular models for typically complex and strongly interacting biological systems are Bayesian Networks.

However, the practical utility of Bayesian Networks is often hampered by the fact that the number of model parameters grows exponentially with the number interactions, leading to rapid overtraining due to the curse of dimensionality.  Here, we propose a solution to this problem by combining Variable-Order Markov Models developed in Information Theory with Bayesian Networks, and we demonstrate the efficacy of the resulting Variable-Order Bayesian Networks for modeling cis-regulatory modules, gene-regulatory networks, and metabolic networks.

30 Min. Presentation + Discussion

12.15: Lunch Break, Discussions, Exhibition

60 Min.

13.15: Topic 3: Dr. Jürgen Sühnel, Leibniz Institute for Age Research - Fritz-Lippmann-Institute, Jena, www.imb-jena.de

Bioinformatics and Systems Biology in Jena: Review and Recent Developments.

Session 2: New Business Perspectives

14.15: Topic 4: Florian Gehrke, Kapelan Bio-Imaging, Leipzig, www.labimage.com, www.kapelan-bioimaging.com

LabImage Platform: A Model Driven Approach for Bio-Imaging Applications

The success of commercial software projects relies on the ability to reach the prototype in a time and cost efficient procedure. Specifying an issue with the help of an abstract Domain-Specific Language clarifies the actual problem and therefore enhances the communication between customer and programmer. Furthermore, the strict separation between model and implementation accelerates the development and minimizes errors and redundancy.
Kapelan picked up essential, recurring patterns in Bio-Imaging applications and translated them into a model-centric software platform. The platform provides many features such as image processing, flexible user interface, operating-system independency and full GLP compliance. This presentation illustrates how the in-house developed LabImage Platform solves the task of rapidly creating quality software for Bio-Imaging problems.

30 Min. Presentation + Discussion

15.15: Topic 5: Dr. Detlef Groth, MPI MP, Potsdam Golm, www-de.mpimp-golm.mpg.de

The Bioscanners project - easy ways to make use of your data.

Data parsing and data of the output of bioinformatic tools is currently not performed in standarized way. To give an example: for the analysis of a common data file like a BLAST output file dozens of individual methods exists. One widely applied approach to solve these tasks is the usage of Bio-toolkits, large frameworks written in a certain programming language. However, this approach has some disadvantages, for the programmer and the end users side: slow parsing speed, tedious and error prone data inspection, and data integration. Therefore, alternative solutions are required. We tested different generators for various programming languages with respect to their ability to produce standalone, small, and fast applications, which can easily be delivered on any modern computer system. The usage of scanner generators ensures a code base, which is small and easy to maintain. With re2c we found a very useful scanner generator fulfilling the requirements described above.

Furthermore applications output database code, which can be imported directly into a SQL-compliant database. As a result, both the installation, usage, and the post-parsing work of data integration and data analysis is greatly simplified. Finally, sophisticated database tools can be used to explore the data interactively. For example our BLASTScanner application is a one single C-file which can be compiled on any modern computer platform to a single 20-50~kB large file, which does not depend on any additional library. The source code for our applications, like the BLASTScanner and binaries for some platforms (Windows, Linux, Mac-OSX, Solaris) is freely available at the sourceforge project page http://bioscanners.sourceforge.net.

30 Min. Presentation + Discussion

Session 3: Clusters & Networking in Bioinformatics

16.15: Topic 6: Martin Pohle, Cluster Life Science Mitteldeutschland, www.medways.eu

The Bioinformatics Networking Platform of Cluster Life Science Central Germany

Since 2005, the Cluster Life Science Mitteldeutschland provides a joint networking platform for agents of biotechnology in the region of Central Germany (Thuringia, Saxony, Saxony-Anhalt). Next to providing extensive cluster and networking structures, the CLCG moves now to a new level by providing an open think-tank structure for the Life Sciences.

With the support of the European Union and the State of Thuringia, CLCG wants to collect and jointly develop new technologies and innovations in bioinformatics into new businesses. This endeavour is open to any bioinformatics researcher and company in Central Germany and beyond. We provide initial funding and our far-reaching business development and cluster network to help create new business models and research endevour. Project Manager Martin Pohle will call for submissions and project ideas in applied bioinformatics provided by the scientific and business community.

30 Min. Presentation + Discussion

17.15: Conclusion and Final Discussion

45 Min. Discussion

18.00: Get Together Dinner + Social Function with all of the participants of the World Conference.