Biology Reference
In-Depth Information
by the LIMS. Without this computer tracking system, the various activi-
ties of the pipeline would remain disconnected from one another. It is the
LIMS that makes it possible to think of a discrete-sample-thing fl owing
through the pipeline; it is only in the computer that the place-to-place
movement of the samples is made sense of, or makes sense. This makes
it problematic to think of the process as a clear-cut transition from ac-
tual sample to virtual sequence inside the detector; as soon as a sample
is barcoded and entered into the LIMS, it becomes a virtual as well as an
actual object. The actual object in the lab would be meaningless without
its virtual counterpart linking it to the pipeline. As the actual-virtual ob-
ject progresses down the pipeline, however, it does become increasingly
liquid—as it is further disconnected from its actual presence, it is trans-
formed into a more mobile form. Images such as fi gure 3.3 suggest the in-
tricate labor that must be performed in the early stages of the pipeline as
the sample is tracked in its minute movements around the lab. By the
fi nishing stage, however, the sample can be effortlessly pulled up onto a
screen and, fi nally, dispatched instantly across hundreds of miles to the
GenBank database in Bethesda, Maryland. The sequencing pipeline can
be understood as a process of enabling DNA to travel.
But the pipeline does more than make things fl ow. The metaphor
also serves to obscure the transformational and judgmental aspects of
the sequencing process. Real pipelines merely transport a uniform sub-
stance from one place to another. The metaphor of a sequencing pipe-
line (rather than a sequencing production line, for instance) suggests
that the extracted text was already inside the body of the sequenced
subject, waiting to be piped out (like oil or natural gas); it suggests that
the As, Gs, Ts, and Cs are merely transported from the organismic body
to hard drives in Bethesda. The pipeline metaphor allows biologists to
understand sequencing as travel through space, rather than as an ac-
tive process of extraction and construction shot through with diffi cult
manual tasks and active judgment calls. Those directly involved in the
day-to-day work of the pipeline rarely experience it as a linear fl ow.
Figure 4.1 shows a process fl ow diagram depicting a far more intricate
set of actions and interconnections than is suggested by a pipe. Organi-
zational and workfl ow charts on the walls of the Broad depicted some-
thing more like a dense network than a pipeline.
The metaphor, however, serves to conveniently collapse the sequenc-
ing process for those who use sequence data; imagining the sequencing
process as a linear fl ow serves to de-problematize the actual-to-virtual
transformation and allows the sense that sequencing is an automatic,
black-boxable activity. Such an elision is crucial in bioinformatics, for
