Introduction - Logic Synthesis for Genetic Diseases - page 6

begin with extracting RNA from cells, converting the RNA to single stranded cDNA
(Complementary DNA), attaching fluorescent labels (markers) to the cDNAs, allow-
ing the cDNAs to attach (bind) to their complementary probes on the slide, washing
the slide of any unbounded molecules, and then detecting the fluorescence of the
attached cDNA on the slide.
The principle of the microarray is that if a specific gene is expressed, then the
corresponding RNA is produced. The RNA is converted to cDNA and the fluorescent-
marked cDNA will attach to its complementary probe on the microarray. Those
fluorescent-marked cDNA that attach to probes will fluoresce (emit light when ex-
cited by a laser) and the intensity of the fluorescence can be recorded as a digital
image (for each probe on the microarray).
Through analysis of the digital image, the intensities of fluorescence reflect RNA
levels, and in turn gene expression levels. The measurements of gene expression
from microarrays can be expressed in the form of ratios or as raw intensity values,
which can be further processed with statistical software [ 7 ] to obtained normalized
or binary expression values. The development of microarrays and other measuring
technologies have allowed for snapshot measurements of the entire genome, driving
research to focus on gene regulation in the complete network, rather than just gene
pair interactions.

1.3 Gene Regulatory Networks

A main focus of genomics is the understanding of the manner in which cells execute
and control the number of operations required for normal cellular function, and
the ways in which cell systems fail, causing disease. While classical approaches in
molecular biology have identified specific processes and interactions in the cell, they
have not been able to produce an overall formalism for cell operation. Many cell
processes, functions, and diseases are a result of highly complex and multivariate
gene interaction, necessitating a system or network view of the genome.
The gene regulatory network (GRN) is one systematic approach to characterize
the cell behavior through gene-to-gene interaction among a set of genes (i.e. how
the expression of a subset of genes affects the expression of another gene in the set).
Several GRN models have been developed, but all models have the same properties,
in that they all represent systems which characterize an interaction among a group
of components as a whole, and they all model a dynamical, time-varying physical
process.
In particular, a GRN model describes the (1) topology (connectivity structure) of
the genes, and (2) the regulating functions of the genes. Both these aspects together
determine the dynamical behavior of system. With an accurate GRN model, an
analysis of the topology and regulation functions can provide deep insight in the long-
term behavior of the system, and identify how the system can fail and lead to disease.
Several models have been proposed for the GRN such as Markov Chains [ 8 ], [ 9 ],
Differential equations [ 10 ], [ 11 ], Boolean Networks (BNs) [ 12 ], [ 13 ], Continuous