Biology Reference
In-Depth Information
scrutinizing visual representations of sequence alignments, they are us-
ing images to reorder, interrogate, and fi nd patterns in their data.
Images also have power as metaphors. Just as texts and codes bring
with them a powerful set of metaphors for biology, images too engender
particular ways of looking at living substance. Because images often
aim to simplify the object they are reconstructing, they perform a kind
of visual synecdoche: the distances between specifi c atoms in a protein
molecule substitute for the complicated three-dimensional structure of
the whole. Or, in the kind of metaphor we have seen here, the genome
or protein becomes a map, a space to be navigated around in one or two
dimensions. Privileging spatial relationships between parts of proteins
or genome sequences is just one way of thinking about these biological
objects. Relationships that are dynamic, temporal, biochemical, or net-
work-oriented are not compatible with seeing the protein or genome as
a static space that can be navigated around. The map metaphor suggests
that biology works only or largely by “navigations”—by fi nding “path-
ways” or “road maps” through the landscapes of genes and proteins.
These maps, trees, charts, and matrices also provide control—they are
ways of rendering the territory of biology as an orderly and disciplined
space that can be manipulated and intervened in.
68
Visualizations are not intended to be “realistic” depictions of bio-
logical objects, but rather to capture particular features of those objects
in order to be able to translate successfully between the computational
and the biological realms. Genomes—as bioinformaticians (and an in-
creasing number of biologists) know them—are necessarily mediated
through these computational representations. It is the structures of the
data that come to constitute the objects themselves. As Brian Cantwell
Smith makes clear, making computational objects is never just an ab-
stract enterprise in computer programming; rather, it is “computation
in the wild”—that is, it is making objects in the world.
69
Genomes are
not totally informatic constructions—their form still refers back to the
molecules from which they derive. But the representational and visual
constructs of bioinformatics—including the database spaces examined
in chapter 5—impose structures and constraints on biological objects;
these structures allow them to interact with certain objects and not oth-
ers, allow them to take certain shapes and not others, allow them to be
manipulated in certain ways and not others.
