Biology Reference
In-Depth Information
complexity of the human genome is likely to be much larger
than that captured even in the richest of all human tran-
scriptome reference sets. Interestingly, only a small fraction
of this previously undetected transcriptional activity appears
to exhibit protein-coding capacity. This suggests that the
complexity of the functional non-coding transcriptome
maybe much larger than the protein-coding one.
Transcriptome variation across individuals within
human populations has also recently been monitored
through RNASeq
[22,82]
, leading to the discovery of
mutations that alter both gene expression and alternative
splicing. Transcriptome variation across individuals
appears not to be as large as variation across cell lines and
tissues
[13]
, and alternative splicing ratios appear to be
more constant across individuals than gene expression.
LncRNAs show higher expression variability than protein-
coding genes. Gonzalez-Porta et al. estimated that about
10% of the human genes exhibit population specific
splicing ratios
[81]
.
to play a role not directly related to protein production.
Characterizing the function of these transcripts, and the
processing pathways through which they are synthesized,
will be the focus of molecular biology for the next few
decades.
Despite technological progress, however, our instru-
ments to monitor the output of transcription
the tran-
e
scriptome
are extraordinarily primitive. There are no
working technologies allowing for the full-length
sequencing of all transcripts present in a given condition,
and methods for transcript reconstruction and quantifica-
tion from short sequence reads are only approximate. High-
throughput techniques to detect and measure protein
abundances are lagging even further behind those for
transcript characterization, as current methods based on
mass spectrometry suffer from both poor sensitivity and
poor specificity. Moreover, although all these methods
generally provide a global overview of the cellular abun-
dances of RNAs and proteins, they have little resolution to
survey subcellular compartments, as experimental proto-
cols to enrich for subcellular populations of RNAs and
proteins tend to be challenging. To fully understand the
function of these RNA species, however, knowledge of
their subcellular localization is essential.
Furthermore, current technologies are in general unable
to interrogate individual cells. Therefore, transcriptome
characterizations emerging from microarray or RNASeq
surveys are very coarse, providing information of the
average behavior of ensembles of tens, hundreds of thou-
sands or even millions of cells. When these ensembles are
relatively uniform
e
The Small RNA Transcriptome
Version 7 of GENCODE reports a total of 8713 annotated
short RNAs. Four main classes represent about 85% of all
of them. Small nuclear RNAs (snRNAs, 1944 genes) are
found within the nucleus of eukaryotic cells. They are
involved in RNA splicing, in regulation of transcription
factors or RNA polymerase II, and in the maintenance of
telomeres. They are always associated with specific
proteins, and the complexes are referred to as small nuclear
ribonucleoproteins. Small nucleolar RNAs (snoRNAs,
1521 genes) are involved in the processing of rRNAs,
tRNAs and snRNAs, and are located in the nucleolus and
Cajal bodies of eukaryotic cells. MicroRNAs (miRNAs,
1756 genes) are post-transcriptional regulators that bind to
complementary sequences on target mRNAs, usually
resulting in translational repression or target degradation
and gene silencing. There are also 624 tRNA genes.
Whereas most of these annotated short RNAs (65%) are
located in intergenic regions, approximately one-third map
within the boundaries of annotated genes.
as in cultures of immortalized cells
e
average behaviors may be confidently extrapolated to the
behaviors of individual cells. However, if interrogating
complex tissues, the behavior of the ensemble may be very
different of that of individual cells
e
which may even be
regulated at the individual cell level, such in neuronal
tissues. Single-cell protocols exist, but monitoring RNA
abundances transcriptome-wide still requires RNA ampli-
fication. Because amplification does not proceed uniformly
across different RNA species, distortions are expected
when the abundances of different transcripts are compared.
Although single-cell single-molecule genome-wide proto-
cols may be in sight, they will still require the destruction of
the biological material. Following the in vivo fate of the
entire transcriptome within an individual cell may remain
forever in the realm of science fiction, but technological
advances in the near future are likely to significantly
increase the resolution at which we can monitor cellular
transcriptomes.
e
CONCLUSIONS AND FUTURE
CHALLENGES
Thanks to recent technological advances we are able to
survey the transcriptional activity of genomes with
unprecedented resolution. The picture that emerges is of
a transcriptional landscape of unanticipated complexity.
The genome appears almost as a continuum of transcription,
where boundaries between genes
REFERENCES
[1] Denoeud F, Kapranov P, Ucla C, Frankish A, Castelo R,
Drenkow J, et al. Prominent use of distal 5
0
transcription start sites
and the very concept of
e
a gene as a defined unit
become increasingly fuzzy.
Many functional RNA species
e
maybe most
appear
e
e
