Biology Reference
In-Depth Information
nected provide theories of biological action. It is now clear that the
sequence of the genome alone does not determine phenotypic traits.
Databases provide ways of linking genomic information to the other
vast amounts of experimental data that deal with transcripts, proteins,
epigenetics, interactions, microRNAs, and so on; each of those links
constitutes a representation of how sequences act and are acted on in
vivo to make life work.
Conclusions
Biological databases impose particular limitations on how biological
objects can be related to one another. In other words, the structure of
a database predetermines the sorts of biological relationships that can
be “discovered.” To use the language of Bowker and Star, the database
“torques,” or twists, objects into particular conformations with respect
to one another.
90
The creation of a database generates a particular and
rigid structure of relationships between biological objects, and these re-
lationships guide biologists in thinking about how living systems work.
The evolution of GenBank from fl at-fi le to relational to federated data-
base paralleled biologists' moves from gene-centric to alignment-centric
to multielement views of biological action. Of course, it was always
possible to use a fl at-fi le database to link sequence elements or to join
protein interaction data to a relational database, but the specifi c struc-
tures and orderings of these database types emphasized particular kinds
of relationships, made them visible and tractable.
One corollary of this argument is that biological databases are a
form of theoretical biology. Theoretical biology has had a fraught his-
tory. In the twentieth and twenty-fi rst centuries, several attempts have
been made to reinvent biology as a theoretical, and in particular a math-
ematical, science. The work of D'Arcy Thompson, C. H. Waddington,
Nicolas Rashevsky, and Ludwig von Bertalanffy has stood out for histo-
rians.
91
The work of all these authors could be understood as an attempt
to discover some general or underlying biological principles from which
the facts of biology (or the conditions of life) might be derived and
deduced. In the twentieth century, such efforts were almost completely
overshadowed by the successes of experimental biology, and molecu-
lar biology in particular. As Evelyn Fox Keller recognizes, however, the
increasing use of computers in biological research has relied on modes
of practice that might be called theoretical. “In molecular analyses of
molecular genetics,” Keller argues, “observed effects are given meaning
through the construction of provisional (and often quite elaborate) mod-
