Biology Reference
In-Depth Information
management system . . . for the sequence data that allows sophisticated
search capabilities comparable to relational data base” were reserved
for the second.
42
Although the NIH intended the two contracts to be
contemporaneous and closely connected, by the time the request for
proposals was fi nally made (near the end of 1981), only the fi rst was to
be funded.
Dayhoff, Goad, and a small group of other computer-savvy biolo-
gists realized that a nucleotide sequence database had to be a sophisti-
cated theoretical apparatus for approaching biological problems. The
majority of their colleagues, however, while realizing the importance of
a repository, believed that making a database was essentially the trivial
process of reading old journal articles and typing in the sequences. The
NIH, refl ecting this latter view, attempted to create a database with this
simple model in mind. For many, the data bank was a “service” and
therefore dubiously worthy of federal support under the aegis of basic
research. Those at the NIGMS who supported the project had to work
hard to generate fi nancial support by stressing the wide range of re-
searchers, including academic, industrial, and medical, who would use
the database for basic research.
43
Moreover, Jordan and her co-workers
promised that the intention of the funding was only to effect a “start-up”
and that it was anticipated that the database would ultimately be sup-
ported by user charges.
44
Like lab apparatus or journal subscriptions,
the biological database was understood to be something that research-
ers could pay for out of their own budgets. While providing support for
basic researchers, it was not an activity that would contribute funda-
mentally to biological understanding.
The NIH issued a request for proposals for a nucleic acid sequence
database on December 1, 1981. Three proposals were forthcoming: one
from Dayhoff and the NBRF, one based on a collaboration between Los
Alamos and IntelliGenetics (a company based in Palo Alto, California,
and run by Stanford biologists and computer scientists), and a further
joint proposal between Los Alamos and Bolt, Beranek and Newman
(BBN) of Cambridge, Massachusetts.
45
On June 30, 1982, the NIGMS
announced that a contract of $3.2 million (over fi ve years) had been
awarded to BBN and Los Alamos. Los Alamos was to be responsible
for collecting sequences from the published record, while BBN was to
use its expertise in computation to translate the data into a format suit-
able for distribution by magnetic tape and over dial-up connections to
the PROPHET computer (an NIH-funded machine based at BBN). The
NBRF was especially disappointed by this decision; others in the com-
