Path: utzoo!attcan!uunet!bionet!fmi.ch!Wallace
From: Wallace@fmi.ch (Andrew Wallace)
Newsgroups: bionet.molbio.bio-matrix
Subject: Response to the recent discussion.
Message-ID: <1482:Wallace@fmi.ch>
Date: 27 May 89 17:35:00 GMT
Sender: daemon@NET.BIO.NET
Lines: 46

Regarding Dave Kristofferson's request for a response from other biologists
about the current bio-matrix discussions, I would like to give my views about
this topic rather than be one of the passive readers.

Since I don't know anything about informatics theory, qualitative modelling or
any of that stuff I cannot give a technical response but I must say that I am
completely blown out by the size of the tasks being contemplated. It seems to
me that a project to collect and represent all biological knowledge would make
the Human Genome Mapping Project appear small by comparison. Indeed, the 
complete human genome would be just a subset of the overall matrix.

This is not to say that such a goal should not be aimed for, it is just that
I and probably many others are concerned with more immediate tasks. For
example, I would really like to see a model or system (call it whatever you
like) which could _accurately_ predict the function of a gene product from
its primary sequence. Maybe this is theoretically impossible, if the required
information is not encoded in the protein or nucleic acid sequence but is
somehow a combination of events from interaction with "chaperone" proteins
or whatever else you can think of to bring about the final functional state.
I don't know how to address this issue, but certainly its a problem which is
relevant to my own research.

Another area which relates to the development of the matrix (which I know
many others have talked about already) is the overall usage of computer-
based information tools by the biological community. I think there are
two, closely related, problems here. One is that the general proficiency
of biologists with computers is not very high compared to other fields.
Most of my (particularly more senior) colleagues barely know how to use
the computer as a word-processor, never mind do a database search, sequence
alignment, access Medline or much else. Secondly the "management" (again
mainly senior colleagues) tend to be indifferent to, suspicious of, or
even openly hostile to using a computer other than to type a letter.
As Dave Kristofferson pointed out however, these problems will tend to
disappear once today's students, who form the major body of computing 
biologists, become the assistant professors and staff scientists of 
tomorrow. Still, this will take time.

To sum up, the matrix project is a great goal to aim for in the long-term,
but if tomorrows computing biologists are to be attracted to it, they must
be convinced that computers can help them to solve more immediate problems,
relevant to their own research needs.

Andrew Wallace, Friedrich Miescher Institute, Basel, Switzerland.

Disclaimer: The opinions expressed above are my own, not those of the FMI
            or its other staff.