Microarray Data Analysis Pipelines - User-Level Workflow Design: A Bioinformatics Perspective - page 116

In order to make a living organism, the genetic information (i.e., the DNA)
has to be interpreted and processed into organic substances. Figure 6.1 (left)
gives a schematic picture of this process, which is also called protein biosyn-
thesis . The first step in this process is called transcription : an enzyme (the
RNA polymerase) produces an RNA copy of a part of DNA. This copy, the
messenger RNA (mRNA), leaves the nucleus and goes into the cytoplasm of
the cell, where it is in the subsequent translation step used as template for the
synthesis of proteins at the ribosomes. The ribosomes “read” the sequences in
triples (called codons ) and translates them into single amino acids according
to the genetic code (right side of the figure). The matching amino acids are
carried to the ribosomes by the transfer RNA (tRNA) molecules and added
to the protein under construction one after another.
Although all cells of an organism have the same genetic material (the
same genotype), they appear in different forms (phenotypes), differing in
physiology and morphology. The reason is that not all genes, but only those
that are currently needed, are transcribed (expressed) and used as RNA or
further translated into proteins, and hence the set of expressed genes (the
transcriptome) varies.

6.1.1 Systems Biology: Transcriptomics

Accordingly, transcriptomics , another sub-discipline of Systems Biology, is
concerned with “global analyses of gene expression with the help of high-
throughput techniques such as DNA microarrays” [279, Glossary]. The fol-
lowing introduction to its principles is largely based on [279, Section 7.1.1].
Transcriptomics aims at studying and understanding the regulation and
expression of genes, which can provide insight into the functions of the gene
products. Typical gene profiling studies are concerned with the comparison
of the gene expression patterns of different cell populations. For instance, the
gene expression profiles of healthy and tumor cells can help to characterize
and classify the tumor properly, which is necessary for finding the optimal
treatment for the patient.
DNA microarray technology allows for high-throughput, time-ecient anal-
ysis of whole transcriptomes of cells. Figure 6.2 gives a schematic overview of
a microarray experiment: DNA microarrays consist of some sort of solid sup-
port material (e.g., a glass slide or a nylon membrane), on which thousands of
nucleic acid spots are arranged close to each other. Each spot contains many
copies of a unique, single-stranded DNA fragment, which can be unambigu-
ously assigned to a specific gene. During hybridization, the RNA that has
been isolated from the cell population under investigation is applied to the
DNA microarray, and the RNA fragments pair with the spots that carry the
complementary single-stranded DNA molecule. After washing (for cleaning
the microarray from of the unbound, left-over RNA), the DNA microarray
is put into a microarray scanner in order to detect the intensity of the spots,
which corresponds to the amount of RNA that has hybridized.