Biology Reference
In-Depth Information
oriented programming language, called Simula, was developed at the
Norwegian Computing Center beginning in 1963 under a contract with
the Sperry Rand Corporation. The need for a language that could be
used to simulate complicated systems derived largely from the fi eld of
operations research.
23
The problem was to design a language that could
easily and accurately describe objects and events in the world:
Programming cannot be started before the system is precisely
described. There is a strong demand for basic concepts useful
in understanding and describing all systems studied by simula-
tion from a common point of view, for standardized notation,
for easy communication between research workers in different
disciplines.
24
In other words, the key to solving complicated simulation problems lay
in representing them inside the computer in a way that matched descrip-
tions in the real world.
Durbin and Thierry-Mieg were faced with a similar problem—that
of representing and relating biological objects (some of them yet undis-
covered or undescribed) inside a computer in a way that would be rec-
ognizable and useful to biologists. Their object-oriented database aimed
to provide a means of describing biological objects from the real world.
In doing so, however, AceDB also placed constraints on what those ob-
jects were. In order to represent the objects, AceDB's designers had to
make decisions about what they
were
—for instance, how would the
class “Gene” be described? What kind of objects could it contain? How
should the hierarchy of classes be structured? These sorts of represen-
tational decisions necessarily produced the meaning of genes, genomes,
exons, and so forth as highly structured objects inside the database.
For Durbin and Thierry-Mieg, storing, sharing, viewing, and under-
standing
C. elegans
data were all part of the one problem. AceDB's vi-
sualization is something that biologists can not only observe, but also
work with, manipulate, and analyze. As such, the ways in which bio-
logical objects are structured within the database have consequences
for how those objects are understood. Because AceDB provides a way of
“seeing” biological objects, the features that it emphasizes become the
features of the object that become “visible” and therefore pertinent to
biological work. Just as a single territory can be rendered as a political
map, a physical map, a topological map, and so on, AceDB produces
“maps” of the genome that show how various objects map onto the
“territory” of the sequence. Under the gaze of AceDB, this territory—
