Genomics Studies - Biomedical Informatics in Translational Research - page 76

research. With the completion of the Human Genome Sequencing Project, the esti-
mate of human protein coding genes is between 20,000 and 25,000. Most of the cur-
rent whole genome microarrays cover almost all of these transcripts. Alternatively,
it is possible to obtain microarrays that focus on specific diseases or pathways, and it
is also possible to have custom arrays prepared for those projects with special
requirements. The major developments will take place in sample preparation, which
will allow the use of smaller samples that have been archived for other purposes.
Such samples were not prepared with RNA analysis in mind and so the RNA quality
will be suboptimal, for example, FFPE samples that are isolated using
microdissected, laser-captured samples. Companies such as NuGEN are focusing on
the amplification of RNA for microarray experiments from as little as 1 ng of total
RNA.
More than 60% of human genes undergo alternative splicing and yield hun-
dreds of thousands of functionally distinct transcript variants. In addition, many
disease-causing point mutations affect RNA splicing. Hence, the need for arrays
capable of detecting these splicing events has increased dramatically in the last few
years. The Affymetrix Exon array is one of the platforms designed to target these
splicing events. It has an increased number of probes that target individual exons of
most gene sequences and so can identify expression levels of individual exons,
which, together with novel analysis algorithms, has allowed the detection of correct
and incorrect splicing events.
Another new and highly focused microarray application for gene expression has
emerged in the last few years that goes beyond traditional analyses. It is known as
genome-wide location analysis and also called ChIP-on-Chip (chromatin
Immunoprecipitation-on-chip). This powerful new platform is used for exploring
transcriptional activities by allowing the determination of the precise location on the
genomic DNA sequence where a regulatory protein is bound. These binding sites
may indicate regions of DNA containing regulatory sites for methylation, histone
modification, DNA replication, modification, and repair and so help in the identifi-
cation of target genes involved in development and disease progression. Such an
approach has been used in research into various diseases such as diabetes, leukemia,
and breast cancer.
5.3.3 Chips for Alternative Splicing Analysis (GeneChip Exon)
The number of genes in the human genome has been estimated by the Human
Genome Sequencing Project as between 20,000 and 25,000, which is considerably
lower than the 100,000 to 150,000 estimated based on expressed sequence tag
(ESTs). Though the number of protein coding genes is less than expected, it is quite
possible that the diversity of proteins made in the human body may well be as high
or higher than the number anticipated from the high number of genes estimated
before the sequencing of the human genome. Alternative RNA splicing is thought to
be the major cause for this diversity. This, of course, suggests that the old paradigm
of “one gene, one protein” (and one function) is an oversimplification.
RNA splicing is a posttranscriptional process that occurs prior to translation. It
is an essential, highly regulated, and complex step. A gene is first transcribed into a
premessenger RNA (pre-mRNA) that contains multiple exons and introns. During