Reliable models SD0030

[gallaher@teleport.com (Ed Gallah]

>most recent issue of Science included several such feedback models of
>dynamics in the immune system.


Thanks for the reply. I know that A.J.P. has been quite supportive of
modeling papers, but I had not seen this particular one. It is definitely
of interest to me, and is close to the type of work I am interested in.

I did see the articles in Science. However, these papers described
differential equations and not SD models. This is not a trivial
difference! SD can be done with an algebra background (along with a
disciplined approach to the modeling process itself, of course).
Practically speaking, anything which uses diff eq eliminates 80% of the
biomedical population. It shouldnt be so, but it really is. After taking
calculus, we receive NO practical application of math to biology in our
remaining undergraduate curriculum. So by the time we get to grad school
math is rusty at best.

Hargrove published a very nice mRNA/protein synthesis model in FASEB
Journal in 1992. I knew it was coming, and then I was disappointed to open
it and find no stock-and-flow diagram. The diagram itself is small, and
very helpful, but they wanted just the results. We need to get those
"funny" diagrams out under peoples noses, at least so they dont look like
hieroglyphics any more.

There are two major points that would change the status quo.

1. Stock-and-flow diagrams, even without simulations, are great for
organizing a discusssion or a lecture. They are not that difficult to
learn.

2. The layperson has no perspective re the construction of a simple SD
model, e.g. linear input, exponential output. (This -very- useful model
should be familiar to everyone. It forms the basis for drug infusion and
metabolism, mRNA synthesis and decay, protein synthesis and decay, Newtons
law of cooling, flushing a reservoir of pesticide, etc., etc.)

This model can be developed in ten minutes by the expert, and in 4-6 hours
by the *absolute* beginner. I had a high school teacher at the end of an
7-hour day asking how he could find the elimination half-time for phenobarb
so he could understand his daughters epilepsy dosage regimen. Others in
the class were asking equally astute questions.

Then I go back to the medical school and this all falls on deaf ears.

Ill need to video one of these teaching days, compress the info into 20-30
minutes, and present this to my colleagues.

Again, Im not saying all this to be whining about it. I know youve all
been through similar situations. However, it just seems so obvious that if
every pre-med student were to take a solid 5-credit SD modeling course in
place of one of the 3 quarters of calculus, wed all be a lot better off.

ed gallaher
gallaher@teleport.com

["Adrian Boucher"]

Hi Ed

I hope you dont mind my picking up the gauntlet which you dropped. I
came across a couple of enlightened SD modellers at one of the UKs
Medical Schools (University of Nottingham), where Neena Washington
taught the medics pharmacokinetics using STELLA on Macs (instead of
the old differential equation method). Results: students began to
understand the problems of drug interactions, understood the
half-life idea, were able to model quite complex dosing regimens,
etc. etc. As ususal, medics had complained about the original
mathematical approach, because they hated the algebra (which they
didnt really need) but needed the understanding (which they didnt
get). The bad news? Neena got moved onto a different course, and
the students now have the pleasure of mathematical pharmacokinetics.
Makes you want to cry, really.

BTW, the Washingtons published a book of pharmacokinetic models:
"Pharmacokinetic modelling using STELLA on the Apple Macintosh". For
once a book whose title says exactly what it contains. Publisher
Ellis-Horwood (I think) or maybe Springer. Publication date: 1992
(?).

Best regards

Adrian

Adrian Boucher
Director
NatWest Financial Literacy Centre
Centre for Education & Industry
University of Warwick
Coventry
CV4 7AL
UK

Tel: +44 1203 524 234
Fax: +44 1203 523 617

e-mail: adrian.boucher@csv.warwick.ac.uk
URL: http://www.warwick.ac.uk/WWW/faculties/ ... index.html

"The e-mail of the species is deadlier than the mail."