I smiled as I read Paul's post. He is right of course. I imagine we
would all agree that modeling is alive and well. There are more models
and modeling papers and modelers every year.
To me the vital question is: Why has modeling failed to gain the
position in biology that it has in, say, physics or engineering?
Sol Snyder, a distinguished neuroscientist and an acquaintance of mine,
could never understand why I wanted to build models. He said, "All the
important discoveries of biology have been qualitative."
What we have failed to make plain to the mainstream, is the power of
modeling to make qualitatively important discoveries. Only when we make
a seemingly outlandish, but genuinely important, assertion that is
subsequently confirmed by experimental test can we hope to make a deep
impression.
Newton and Einstein passed this test so many times that Physics was
forced to embrace theory.
I think polygenic diseases, like cancer and atherosclerosis, cannot be
solved without modeling. When the single gene hopes are dashed on the
rocks of brutal experiment as has been the case for many knockout
experiments (a dramatic one in this week's Science), the community will
turn to double gene knockouts. These won't work either, and they will
then come face to face with the awful truth (well known by the cognitive
psychologists) that the human mind can keep track of no more than three
interacting variables.
We should be developing the large-scale tools we will need when they
arrive on our doorsteps. When theory shows the way to a cure for cancer
or for heart disease, then biology will embrace theory, and someone or
some team will enter the history books as "the Einstein(s) of biology".
Bob climbs down from his soapbox and wanders off....
--
Robert D. Phair, Ph.D. rphair at ix.netcom.com
BioInformatics Services http://www.webcom.com/rphair
Partnering and Outsourcing for Computational Biology
