Biology Reference
In-Depth Information
which “the sequence” is the central object. Understanding sequencing
as a directed, unproblematic process—a simple linear motion through
space—allows the output products (the digital renderings of the se-
quences themselves) to be plausibly used as proxies for the input mate-
rials (wet biological samples). Almost all of bioinformatics depends on
this fact that sequences can be relied on as straightforward extractions
of biological material into biological data. This section's detailed de-
scription of the sequencing pipeline shows, however, that the produc-
tion of biological knowledge (in this case, sequence) depends crucially
on carefully contrived movement through space. Such movement is far
from automatic or linear—it is shot through with judgment calls and
contingent processes. The production of “fl uid” and “universal” bio-
informatic objects depends on a highly situated process through which
this “fl uidity” and “universality” is constructed; bioinformation can
travel and fl ow only because its movement through the pipeline serves
to obscure its solidity and situatedness.
The pipeline metaphor is not intended to completely capture the
features of a biological specimen. Indeed, such a task would be impos-
sible. Rather, the actual-virtual transition reduces the sample to a digital
trace, fl attens it into a text that can be computed, communicated, and
manipulated in virtual space. Sequence data, consisting of a base call
and a quality score, already constitute a structure to which biological
sequences must conform. Following the data to their place of produc-
tion shows how they are constructed for a particular purpose; they are
crafted as data objects ready to fl ow through computers and networks.
Standards and Ontologies
Simply producing sequence data and putting them in GenBank would
not be useful for biological research unless bioinformaticians and biolo-
gists already agreed about standards and formats for writing, storing,
and reading such data. GenBank is doing far more than just storing
biological data—it is enacting common standards for the reading, writ-
ing, and storage of biological information. This section will examine
how bioinformaticians go about developing and enforcing common
languages. Such languages are required not only for communication,
but also for making knowledge. When language is fl attened into data,
it becomes possible to create ways of agreeing about objects and their
proper descriptions.
For nucleic acid sequences themselves, a standard shorthand was al-
ready in place: A, G, T, C, and U were in use before large-scale sequenc-
