Biology Reference
In-Depth Information
microarray technology has been extensively used to compare or quantify genome-
wide mRNA levels in both eukaryotic and prokaryotic organisms.
The first purpose of a transcriptome study is often to identify genes that are
differentially expressed between two or more conditions, and statistical testing
for differential expression of genes is now a relatively streamlined procedure. How-
ever, making biological sense of lists of differentially expressed genes is often not
trivial and many commonly applied analytical approaches extend beyond testing for
the differential expression of single genes. This includes, for example, the analysis of
the expression patterns of predefined gene sets (groups of functionally related genes;
reviewed in
Ackermann and Strimmer, 2009
) or the application of clustering
methods whose purpose is to delineate sets of genes that share similar expression
patterns across the experimental conditions and thereby facilitating the tentative
assignment of functions to previously uncharacterized genes (reviewed by
Quackenbush, 2001
). Furthermore, the availability of data sets collected in different
genetic backgrounds and under a number of experimental conditions initiated the
development of methods to infer transcriptional regulatory networks, most often
combining cluster analysis of gene expression data with transcription factor binding
motif searches (reviewed by
Herrg
˚
rd
et al.
, 2004
).
Recent technical advances in high-density array formats and DNA sequencing
technologies (i.e. next generation sequencing, NGS) associated with the growing
number of sequenced bacterial genomes have opened the way to the characterization
of entire transcriptomes by genomic tiling arrays and whole-transcriptome sequenc-
ing (RNA-seq) methods. The rapid development of these powerful tools for unbiased
transcriptome analysis is reflected in several recent reviews on prokaryotic transcrip-
tomics (e.g.
Sorek and Cossart, 2010; M¨der
et al.
, 2011; Filiatrault, 2011
). Genomic
tiling arrays, which are high-density oligonucleotide arrays covering both strands of
a genome by overlapping probes, facilitate the mapping and quantification of all tran-
scribed regions with a resolution determined by the distance between two adjacent
probes which is typically 10-20 nucleotides in the case of microorganisms. Whole-
transcriptome studies based on tiling arrays and/or RNA-seq have revealed an unex-
pectedly high level of complexity within the bacterial transcriptome (
Sorek and
Cossart, 2010
), even when only a relatively small number of experimental conditions
are analysed. A recent tiling array study of
B. subtilis
grown under a wide range of
nutritional and environmental conditions has established a comprehensive repertoire
of transcription units and identified, for example, more than 1500 new RNA features
including 512 potentially new genes (
Nicolas
et al.
, 2012
). For the first time, this
study evaluated the contribution of a bacterium's RNA polymerase sigma factors
to global transcriptional regulation, an analysis made possible by the availability
of promoter-level data with high signal-to-noise ratio.
Systems biology approaches require the comprehensive identification and
precise quantification of all transcripts of a given sample. In line with this, genomic
tiling arrays can be used to detect and annotate novel transcripts including 5
0
- and
3
0
-untranslated regions, protein-encoding mRNAs as well as regulatory and catalytic
RNAs,
thereby expanding the existing genome annotation. Furthermore,
they
