Information Technology Reference
In-Depth Information
4 Modeling Genetic Networks
4.1 Why Modeling?
Most of diseases are caused by a set of gene defects, which occur in a complex
association. The association scheme of expressed genes can be modelled by genetic
networks. The key is to keep things simple, at least to start with
[18].
Just measuring
levels of mRNA tells scientists that a gene has been activated, but does not detail the
amount of protein it encodes, or which task that protein fulfils in cell dynamic. The
future will be the study of the genes and proteins of organisms in the context of their
informational networks. Scientists from independent Institute for Systems Biology
(
ISB) plan to produce a complete mathematical description of complex biological
systems e.g. the immune system and complex conditions such as cancer or heart
disease. All these efforts will culminate in the assembling of the biological equivalent
of a virtual cell. Genetic networks are efficient facilities to understand the dynamic of
pathogenic processes by modelling molecular reality of cell conditions.
Genetic networks consists of first, a set of genes of specified cells, tissues or species
and second, causal relations between these genes determining the functional condition
of the biological system, i.e. under disease. A relation between two genes will exist if
they both are directly or indirectly associated with disease. Our goal is to characterize
diseases - especially autoimmune diseases like chronic pancreatitis (CP), multiple
sclerosis (MS), rheumatoid arthritis (RA) - by genetic networks generated by a
computer system. We want to introduce this practice as a bioinformatic approach for
finding targets.
Genes that follow similar patterns of expression are likely to share common molecular
control processes. Furthermore, assuming that there is a reason for these genes to be
expressed together, it is possible that they participate in a similar or complementary
set of functions required by the organism under a given condition
[19].
Comparison of
expression profiles will not deliver the kind of intimate understanding of the highly
inter-related control circuitry that is necessary to achieve true understanding of
genome function
[5]
. But with an optimal composition of expression profile analysis
methods mentioned above combined with modelling we will get a powerful
instrument to describe diseases in a highly abstract way. So we have to combine the
library of tools we use to analyse expression data - recruiting as well statisticians and
mathematicians as biologists and physicians to consider multi variant problems of a
never knowing size and complexity. In the next few paragraphs methods are discussed
which we want to combine for generating genetic networks.
4.2 Processing an Artificial Neural Network to Classify Gene Expression
Patterns
An artificial neural network is utilized for classifying diseases to specific diagnostic
categories based on their gene expression signatures. We chose a neural network of
adaptive resonance theory (ART). An ART net works like a self-organizing neural
pattern recognition machine. The five major properties of the ART system are
plasticity as well as stability, furthermore sensitivity to novelty, attentional
