Transcript BMBF-Funding Initiative “Systems of Life – Systems Biology” Platform Modeling / Bioinformatics Coordinator: Prof.

BMBF-Funding Initiative “Systems of Life – Systems Biology”
Platform
Modeling / Bioinformatics
Coordinator: Prof. E. D. Gilles
Presentation Heidelberg, July 7th 2004
Sven Sahle, EML research gGmbH
Platform partners
Humboldt University, Berlin:
Prof. H.-G. Holzhütter, Charité, Mathematical Modeling
Prof. R. Heinrich, Biology, Theoretical Biophysics
Prof. T. Höfer, Institute for Theoretical Biohysics
Prof. A. Herrmann, Biology, Molecular Biophysics
Prof. H. Herzel, Institute for Theoretical Biology
Prof. J. Reich, MDC for Molecular Medicine,
Bioinformatics
EML Research, Heidelberg:
Dr. U. Kummer, Bioinformatics and Computational
Biochemistry
Dr. R. Wade, Molecular and Cellular Modeling
MPI DCTS, Magdeburg:
Prof. E.D. Gilles, Systems Biology
Prof. S. Schuster, Univ. Jena, Bioinformatics
Mission:
The platform “modeling/bioinformatics” is devoted to the
development of methods and tools for the efficient construction,
analysis, integration and exchange of complex mathematical
models in systems biology.
The platform interacts with all partners of the initiative by:
•Providing novel methods and tools for the systems-level analysis
of the hepatocyte.
•Conducting specific research projects in cooperation with the
partners to develop methods, tools and standards.
Key objectives of research and development
• Unified methodology for the kinetic modeling of complex cellular
networks encompassing metabolic, signal transduction and
genetic sub-structures with a focus on network representation
and complexity reduction.
• Novel methods for the analysis of complex networks based on
systems theory regarding structural properties, network
decomposition, identification of model structures, and others.
• New and/or improved computer tools for standardized modeling
and simulation, including model and data storage.
• Integration of experimentation and modeling with respect to
efficient experimental design and real-time control of biological
processes.
Coordination of activities
• Distribution of tasks both in development of mehods/tools and
research on cellular systems
• Progress meetings of the platform partners every 6 months.
• Annual international workshop `Modeling and simulation of
complex biological systems´ open to all researchers within the
BMBF initiative.
• Internal web portal for modeling and bioinformatics
Kinetic modeling of complex cellular
networks with special focus on
hepatocytes
I. Methods and tools
•
Generalized control theory of cellular networks based on the wellestablished concept of metabolic control theory (Heinrich/Kacser).
•
Standards for the formal and graphical representation of cellular
networks.
•
Theoretical framework for identification and evaluation of potential
interfaces between various types of cellular networks.
•
Inter-active software modules for computer simulations of hepatocyterelevant kinetic models.
II. Modeling of selected sub-networks
…
Kinetic modeling of selected sub-networks:
successive development of an integrated model
integrated kinetic model
Camediated
cell-cell
interaction
Wnt-ßcatenine
signaling
pathway
platform cell
biology „3D
bioartificial human
liver cell systems“
(Berlin/Jena)
expression
control of
metastasis
genes
ubiquitindependent
protein
turnover
metabolism
and
biogenesis of
lipoproteins
cooperation
applications
(examples)
network project
„systems biology
of primary and
regenerating
hepatocytes
(Freiburg)
project
„vectorial
transport
through
virtual
hepatocytes”
(Heidelberg)
intracellular
lipid traffic
interfaces between the various modules of the integrative cell model
identification of potential
oncogenes and tumor
suppressor genes in
signaling pathways
prediction of systemic
effects upon administration
of proteasome inhibitors
identification of target
enzymes for the
pharmacological
treatment of disorders in
the lipid metabolism of
the liver
Characterization of complex signaling and regulatory
processes in hepatocytes using modeling and systems
theory analysis
I. Methods and tools
•
Modeling concepts for regulatory networks
•
Visualization of models and simulations in ProMoT
•
Structural analysis of signal transduction networks
•
Software sensors for process control
II. Model-based analysis of selected sub-networks
•
Mitogenic and apoptotic signaling pathways
•
Signal integration in proliferation control
Characterization of complex signaling and regulatory
processes in hepatocytes using modeling and systems
theory analysis
SYCAMORE
I. Evaluate and integrate existing methods
II. Develop new methods
•
Complexity reduction of big models
•
Hybrid simulation methods
•
Structure based methods to compute kinetic constants
•
Sensitivity analysis of higher order
•
Semi-automatic generation of models from databases
III. Apply tools to selected sub-systems of the hepatocyte
Cellular systems
Methods and tools
Modeling
methods
Reduction of complex
kinetic models
Further development of
standards for model
exchange (SBML)
Symbolic
representation of
elementary processes
and networks
Network
analysis
Computer-based
tools
Identification and
evaluation of interfaces
between cellular
networks
SYCAMORE
expert system for
mathematical modeling
and experimental
design
Generalized control
theory for cellular
networks
Models and
experiments
Parameter estimation
from system data and
protein structures
Software sensors for
hepatocyte bioreactors
PROMOT/DIVA
modeling / model
library, simulation,
model analysis
Structural analysis of
regulatory networks
Model-based
experimental design
Interactive software
modules for computer
simulation
Metabolic networks
Metabolism and biogenesis of
lipoproteins
Signaling networks
Gene networks
Wnt/ß-catenine pathway
Cell cycle regulation
Ca-mediated cell-cell interactions
Expression control of metastasis
genes
Intracellular lipid transport
Ca-mediated cellular signal
transduction
Cytochrome P450 enzyme
systems
Ubiquitin-dependent protein
turnover
Mitogenic and apoptotic pathways
Heidelberg
Magdeburg
Berlin
All groups
COPASI
a simulator for complex
pathways
modellin
g
simulation
analysis
reporting
Traditional tools:
modellin
g
simulation
analysis
reporting
command line tool
text editor
command line tool
plotting tool
(eg. gnuplot)
COPASI will combine all this in one tool with a graphical user
interface. Users of COPASI should be biochemists and biologists
without expert knowledge about simulation methods.
-> promote methods of systems biology
modellin
g
simulation
analysis
reporting
How is the biochemical reaction network described in
COPASI?
• there are some chemical species
• species are involved in chemical reactions
• reactions happen with a certain speed.
• all this happens in a compartment (of the cell)
Compartments just have a Volume.
Species are contained in compartments. They have a
concentration or particle number (which can be converted
using the volume of the compartment)
modellin
g
simulation
analysis
reporting
Deterministic simulation
The model is converted to a set of differential equations.
The simple example (A -> B, v = k*substrate/(kM+substrate))
will give:
dA/dt = -k*A(kM+A)
dB/dt = +k*A(kM+A)
These differential equations are then numerically integrated
using the LSODA solver (Adams for nonstiff regions, Gear
for stiff regions).
Some Details
• written in C++ using QT library
• available for Linux, Unixes, MacOs X, and Windows
• will be free for academic use
COPASI is developed in cooperation with Pedro Mendes,
Virginia Bioinformatics Institute, Blacksburg, USA
Conclusion:
COPASI will be an easy to use tool including powerful
standard methods of systems biology.
COPASI also acts as a framework for the new modelling,
simulation, and analysis tools that are developed in the
BCB group
SYCAMORE
SYCAMORE (Systems biology Computational Analysis
and MOdelling Research Environment) is a project
carried out at EML Research, Heidelberg with the
following aims:
• Build a suite of methods and tools to faciliate the
integration of experimental and computational
approaches
• Support the user in the choice of appropriate
computational tools to tackle a specific problem
SYCAMORE
In order to develop SYCAMORE we need to
• Evaluate and integrate existing methods and
• Develop new methods in
• Complexity reduction of big models
• Hybrid simulation methods
• Structure based methods to compute kinetic constants
• Sensitivity analysis of higher order
• Semi-automatic generation of models from databases
SYCAMORE
SYCAMORE architecture: