Gene regulatory networks (GRNs) represent the dependencies of the different actors in a cell
operating at the genetic level. They dynamically determine the level of gene expression for each gene
in the genome by controlling whether a gene will be transcribed into RNA. A simple GRN consists of one
or more input signalling pathways, several target genes, and the RNA
and proteins produced from those target genes. In addition, such networks often include dynamic
feedback loops that provide for further regulation of network regulation activities and output.
In order to understand the underlying structures of activities and interactions of intra-cellular processes
one has to understand the dependencies of gene products and their impact on the expression of other genes.
Therefore, finding a GRN for a specific biological process would explain this process from a logical point of view.
JCell is a framework for simulating GRNs. It is completely implemented in Java and can be used for two different applications:
reverse-engineering and inferring regulatory mechanisms based on the evaluation of given biological and medical
data coming from DNA microarray experiments, and
simulating cell growth and mitosis by finding GRNs suitable for a given problem (e.g. limited growth).
Gene regulatory network analysis exploits massively parallel measurements of interacting biochemicals, namely with
DNA microarray techniques. The measurements at different states of the cell can be used for studying the
relationships between each component of cellular processes by mathematical modelling of the dependencies in the data set.
There are different types of mathematical methods implemented in JCell for simulating GRNs like
The parameters of each model are evaluated either by optimization with
Evolutionary Algorithms or
by straight-forward heuristics, if available.
- Random boolean networks (RBN),
- Weight matrices,
- pseudolinear Weight matrices,
- S-Systems, and
- arbitrary differenatial equations
The main focus of the current research is on GRNs related to immune-specific diseases, as JCell is developed
as part of the TuebinGENome project of the NGFN
(Nationales Genom-Forschungsnetz) in cooperation with the
Cell Simulation/ Morphogenesis
The translation of the dynamics of gene regulatory networks in terms of pattern and form is a central problem,
not only for biology, but also for Artificial Life.
The simulation of cells or organisms could be used to study the effects of genetical defects or gene-related diseases or
the impacts of drugs.
© 2004 University of Tübingen, Germany