Managing a molecular biology computer resource

JRAMON at es.uam.fmed.mvax JRAMON at es.uam.fmed.mvax
Thu Aug 6 11:56:00 EST 1992


In view of this recent discusion and some of the answers, I'd like to post
my own opinion on the subject.

	I came to this field from Molecular Biology, and to that from Medicine,
and I am now directing the Computing Facility here. I can't think of it as a
good pathway. Neither can I think it would be OK for a biologist or else.

	First, Biology (& Medicine, & Pharmacology, & etc...) is/are evolving
VERY rapidly now. The same is true for Computer Science. However, most
curricula nowadays don't cope well with more than one of these disciplines
(nor they should). In this respect, any of these people (or engineers, or
even plomberers) are equally far of a Biocomputing Manager.

	Second, IT IS NOT BIOCOMPUTING!!! It is medical, pharmacological,
chemical, physical, even mecano-quantical computing as well. It DOESN'T make
sense to propose any special career as th best. What is needed is a general
ability to understand and analyze problems (preferrably with scientific
methodology), and !!! an open predisposition to this field !!!. This way,
any person that has been able to earn a universitary degree and is willing,
can learn anything new he/she needs.

	Third, and again, you MUST wish to learn and be open-minded. This
field requires deep incursions in many different fields. You need to
UNDERSTAND the algorithms, how they work, what they do and why. BUT I can't
accept that programming is a must. And you must be able toUNDERSTAND what
the needs of other scientific colleages are. Remember, these are all
fast-changing fields. It doesn't matter what you know. In one or two years
you'll be out of many areas. But if you want, you can learn about them if
necessary (that's what ANY scientist is doing always).

	Fourth, obviously you have to learn about computers. What? I've seen
the revolution go from biocomputing in the Apple II to supercomputing, and
many of you have seen even more. I have handled more than 10 distinct
operating systems, more than 20 different machines, from the Apple II to
the Cray, and each was different. What to learn then?

	If you are able to understand the algorithms, what they do and why,
you can understand WHEN and HOW to use them (or the programs) appropriately.
That's all you need (though you could do many other thnings, like
programming). As for actually HANDLING these resources, a general knowledge
of how the usual computer works is enough. To acquire this knowledge, it
could suffice learning two different systems (one of them at least
command-oriented, not graphical). By comparing you'll soon learn what the
basic operations are. And that's all you need.

	Later, when you face any new system, problem, algorithm, etc... with
an open mind, a bit of intelligence and a willing mood, it shouldn't be much
of a problem.

	Fifth, and most important, you need to be able to integrate finely
in an adverse environment, and to collaborate with your worst enemies to
help them do their work. Then you'll be able to help all the other guys
in your center. Then you'll really be a professional.

	As for the deepness of the knowledge each one must decide how much
time is willing to invert in this area. A scientist, may decide to learn
the basics of user-level for a generic computer (he'll work surely with
more than one), the most basic of computer logic (to understand a simple
explanation of an algorithm he needs), and a few of good-manners. This doesn't
need much time.

	If you are to make a living of this, you may as well invert two or
more years to learn the basics of system management, system analysis and
program design. You don't need to be a programmer, just an advanced user
level could be enough. But you'd better continue always learning all your
life.

	And in between, there may be all the colours of intermediate interests.


	Final note: I'd been working on the curriculum for Biochemistry
studies. My best bet was that two courses, one to teach the basics of
computer use, and the other to understand how computers and 'biological'
algorithms work was more than enough. If you really needed Biocomputing
scientists, you'd look for them in Computer Sciences. At last, it is
about ANY computing (included bio-) that C Sci is all about.


	Sorry for the extension, but I thought that it was time to deveal
the mistery of "computer gurus" (which really aren't), and demistify it
a bit. At the end all is a function of the time you want to invest and
the results you want to get. Anyone can clean his/her own lab, but most
will leave it to other professionals, or either they would lose most
of their time cleaning instead of researching. The same holds for ALL
areas of work.

		Jose R. Valverde
	Biomedical Research Institute
		Madrid - SPAIN




More information about the Bio-soft mailing list