Biology Reference
In-Depth Information
In a description of an updated version of his programs published in
1984, Ostell made a bold claim for the value of his work:
Adequate understanding of the extensive DNA and protein se-
quence derived by current techniques requires the use of com-
puters. Thus, properly designed sequence analysis programs
are as important to the molecular biologist as are experimental
techniques.
66
Not everyone shared his view, including some members of Ostell's PhD
committee at Harvard. “It wasn't something that biologists should be
doing,” according to Ostell's recollection of the reaction of some mem-
bers of his committee.
67
Despite Kafatos's support, Ostell was not per-
mitted to graduate on the basis of his computational work. Ostell's pro-
grams had made a direct contribution to the solution of many biological
problems, but the software itself was not understood to be “doing bi-
ology.” Even in Ostell's own writing about his work, he describes the
functions of his programs as “data management” and “data analysis,”
rather than biology proper.
68
Ostell could not get his PhD, but Kafatos agreed to allow him to stay
on as a graduate student provided that he could support himself. He
got permission to teach a class called “Computer Anatomy and Physi-
ology” to undergraduates. This class was an introduction to computer
hardware that analyzed the machine as if it were a living organism.
In 1984, Ostell was approached by International Biotechnologies, Inc.
(IBI, a company that had been selling restriction enzymes and labora-
tory equipment), which wanted to license his software and develop it
into a product. Since Ostell had done the work while a graduate student
at Harvard, the university had a legal claim to the intellectual property
rights. But it saw no commercial value in Ostell's work and agreed to
sign over all rights. Turning the software into a commercial product
was a formidable task. In particular, the software had to be carefully
reengineered for reliability, compatibility, and interoperability. The IBI/
Pustell Sequence Analysis Package was released in August 1984, ready
to use on an IBM personal computer, at a cost of $800 for academic and
commercial users. Still unable to graduate, Ostell followed his wife's
medical career to Vermont, where he lived in a nineteenth-century farm-
house and adapted his programs for use, fi rst on MS-DOS and Unix
machines and then on the new Apple Macintosh computers (the latter
version eventually became the MacVector software).
