Biology Reference
In-Depth Information
FIGURE 2.2 Typical illustration of the growth of sequence data. Historical trends in computer storage
prices are plotted against DNA sequencing costs. This is a logarithmic plot, meaning that the straight
lines correspond to exponential growth. The cost of computer storage (squares) is shown as growing
exponentially with a doubling time of roughly 1.5 years (equivalent to Moore's law). The cost of DNA se-
quencing (triangles) is shown in two separate lines corresponding to pre- and post-NGS (next-generation
sequencing). For the latter, the doubling time reduces to six months. (Stein, “The case for cloud comput-
ing.” Reproduced from Genome Biology under the BioMed Central open access license agreement.)
machines, heckling their directors and department chairs to authorize
the expenditure (about $350,000 for a single machine plus about
$10,000 a week for the chemicals to keep it running continuously) lest
they be unable to continue to do biology at the cutting edge. 17
But this need for quantity extended from the technology into the
work itself. During my visit, the Burge lab was devoted to learning about
the nature of RNA splicing. In particular, the group wished to discover
how RNA splicing was regulated and what role this regulation played
in protein function. 18 What was interesting about the Burge lab's work,
however, was that it hardly concerned itself at all with particular splic-
ing events. Instead, the questions the team wished to answer concerned
the nature of all splicing events. How many genes in the genome are
spliced in a particular way? How often? How many ways can they be
spliced on average? Are there any patterns in which exons get skipped?
What can we learn about the ways splicing is regulated from the total-
ity of events? 19 Answers to such questions require an immense amount
of data—the Burge lab needed multiple sequences of the same gene ex-
tracted from different tissues at different times, under a variety of condi-
tions. Only next-gen machines could make such data available.
Search WWH ::




Custom Search