Body Expression Map of Human Genome
79
functions. Altogether, combined with ge-
nomic sequences, we will understand the
exact mechanisms and signals of gene ex-
pression control written in genomic DNA
in the near future through experimental
data obtained from such studies.
1.2
Body-Map Data
Much of the EST approach, such as the
gene cataloguing project, is qualitative. In
contrast, body-mapping projects are de-
signed to clarify the differences of cells
or tissues in terms of gene expression
p
roF
l
e
s
;thu
s
,th
e
yh
a
v
etob
equ
an
t
i
t
a
-
tive at every step during the experiment.
In general, the following points have to
be considered very carefully when carry-
ing out an expression-proFling experiment
or interpreting the experimental data ob-
tained from such projects. ±irst, it is
important to know the origin and qual-
ity of the tissue or cells from which
cDNA libraries are prepared. This is es-
pecially important when comparing data
obtained from different mRNA sources
such as tissues, organs, individuals, or
species. Secondly, the quality and com-
plexity of the cDNA library have to be
carefully examined. Thirdly, any sampling
procedure potentially has biases that will
severely affect the composition of the F-
nal dataset. Without these precautions and
evaluation of starting materials and ex-
perimental steps, the data will lead to
improper conclusions. Statistical assess-
ment of the data as well as the linearity of
measuring technology are also important,
especially when using fluorescently la-
beled material because both body mapping
and DNA chip technology have experi-
mental characteristics that users have to
be aware of when interpreting the Fnal
data.
2
Integration of Millions of EST Alignments
on the Human Genome
Before the human draft genome became
available, ESTs were classiFed primarily
on the basis of sequence similarities. Uni-
Gene, for example, employs this method
to produce clusters of ESTs. In contrast,
the advent of the draft human genome se-
quence makes it possible to align millions
of ESTs with the genomic sequence.
Integration of these alignments is help-
ful in identifying groups of ESTs that
are coded at the same locus, in gather-
ing information on alternatively spliced
transcripts and their representatives (see
±ig. 2), and in associating gene expres-
sion patterns with speciFc loci. It is
noteworthy that these two approaches are
likely to output different clusters for ESTs
that have low-level sequence similarities.
This is because the traditional sequence-
similarity approach tends to overlook a
typical cluster of alignments that share
a small number of exons at the same
positions.
2.1
Association of Gene Expression Patterns
with EST Groups on the Human Genome
Correlation of expression patterns with
loci of gene clusters is a crucial step
in functional analysis. In an effort to
achieve this task, one can attempt to
align ESTs associated with gene expression
patterns in various tissues with the human
genome. ±or instance, the expression
levels of BodyMap representative genes in
30 distinct human tissues are available.
BodyMap sequences are
hundreds
of
base pairs in length and are therefore
sufFciently long to locate their positions in
the human genome. This makes it possible
previous page 753 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online next page 755 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online Home Toggle text on/off