Magdeburg Centre for Systems Biology (MaCS)
3.10. - 5.10.2010
Petri nets provide a powerful graphically and intuitively understandable unifying modeling framework for qualitative, stochastic and continuous analysis & simulation as well as reverse engineering. The course specifically addresses both dry-lab and mathematically naive wet-lab systems biologists and exemplifies the specific benefits of Petri nets for the research of both groups. The course is also open for graduate and PhD students.
Brandenburg University of Technology at Cottbus, Computer Science Institute, Germany
monika.heiner (at) informatik.tu-cottbus.de
Otto von Guericke University Magdeburg, Germany
marwan (at) mpi-magdeburg.mpg.de
All talks are in English.
|day 1||evening||welcome dinner|
|day 2||morning||basics of Petri net theory + practical part|
|afternoon||stochastic Petri nets and colored Petri nets + practical part|
|day 3||morning||analysis and model checking + practical part|
|afternoon||network reconstruction + practical part|
The first lectures of the summer school focus on modeling biological systems in terms of Petri nets and analyzing processes within this framework. The last part of the summer school is dedicated to the challenging task of reconstructing such models from experimental data.
The current state of the art approach tries to cope with the reconstruction problem by cycling through several phases: forming hypothesis, model construction, model-based analysis/simulation, and wet-lab validation of the findings. This approach is hypothesis-driven and yields potentially incomplete information, as e.g. no certificate can be given that no suitable model alternatives exist.
Our group developed an exact mathematical approach that is exclusively data-driven and generates the complete list of all minimal place/transition Petri nets fitting the experimental data. Hereby, the set of biological species considered to be crucial for the observed phenomenon is part of the input and will form the set of places in each of the reconstructed networks. The task is to find all minimal sets of transitions which are able to reproduce the experimentally observed time-dependent mass or signal flux in the system.
In this course, we present the main features and main steps of the reconstruction algorithm:
1. The input: Which type and quality of data is suitable
2. The core routine of the algorithm: How to analyze the experimentally observed state changes of the system in order to determine suitable transition sets of the networks
3. When the algorithm returns no solution: How to detect and resolve infeasibility by extending the network
4. When the algorithm returns too many solutions: How to exclude unsuitable solution alternatives by considering advanced dynamic aspects or by designing new experiments
The course will be accompanied by a practical part, including exercises to step-by-step reconstruct Petri nets from given experimental data and a demonstration how to run the algorithm on some biological examples.
Further information concerning the scientific programme, the list of referents as well as an overview for the planned social activities are following.
summerschool-macs (at) ovgu.de
Otto-von-Guericke University Magdeburg: http://www.uni-magdeburg.de/
Max Planck Institute for Dynamics of Complex Technical Systems: http://www.mpi-magdeburg.mpg.de/
If you need a hotel booking please contact Janine Stierwald stierwald (at) mpi-magdeburg.mpg.de.