MCA Analysis or Theory?
I can't remember now why Henirk Kacser preferred to call MCA an analysis
rather than a theory, perhaps it was to emphasize that MCA was an analysis
approach, that is given certain postulates about how one thought metabolic
systems behaved, particularly the components, one could 'work out - analyse'
the consequences this had on the behaviour of the entire system, a very
powerful trick if one could pull it off.
In the late 80s I did consider MCA a theory (see theorems in FAQ), it seemed
to have all the right bits: a set of basic observations cast as mathematical
axioms (not suggesting that theories have to be mathematically based); a
whole range of consequences ('theorems') and expected patterns of behaviour
derived from the axioms; and most importantly the theorems were falsifiable
through experiment, at least in principle. However at the time and
particularly even earlier in the development of MCA, the word theory was
quite a 'dirty' word in mol biol/biochem circles. I remember as an
undergrad, our metabolic pathway lecturer ridiculing the so-called 'armchair
biochemists' (i.e the theorists) and that experiment alone was the only true
way forward for our science (perhaps unknown to him he was of course
theorizing all the time if he'd only stop to think!). I grew up, so to
speak, and I'm sure many others did too, thinking that theory was something
that one didn't speak too loudly about, and computers, well one didn't go
near computers! At the end of my undergrad days I was 'advised' not to
pursue a theory based research career, which is what I wanted, but should
instead concentrate on one based on experiment.
MCA, for me at least, became MCA for political reasons and MCT was quietly
dropped from my vocabulary. However, things are changing and seem to be
changing very rapidly. With the sequencing of the human genome, for some
unknown reason (god knows why it didn't occur to them before), the molecular
biologists are *now* asking whole system questions, or rather their saying
that we should be looking at the system as a whole and that we'll need
computers etc to sort it all out, they're not quite sure how to go about it
but they're aware that there is a problem. I've seen numerous articles in
the last 12 months or so in very 'respectable' journals proclaiming the new
way forward, for example I have in front of me:
Molecular Interaction Map of the Mammalian Cell etc, in Mol Biol Cell, vol
10 1999, Kurt Kohn, to quote from the first line in the abstract:
'Eventually to understand the integrated function of the cell cycle
regulatory network, we must organize the known interactions in the form of a
diagram....' etc.
And *even* Nature in Dec 1999, vol 402 Hartwell et al has a review titled,
'From molecular to modular cell biology'.
and there are others, even among the popular press such as the New Scientist
and this month in Time magazine.
So perhaps I/we should now call MCA, MCT?
Herbert Sauro
"David Fell" <daf at brookes.ac.uk> wrote in message
news:4.3.2.7.0.20000818142954.00b05290 at csmail.brookes.ac.uk...
> I'm sorry I was away on holiday when this discussion faded out because I
> think it deals with a crucial point that merits continued
> consideration. For a long time I was happy to accept Henrik Kacser's view
> that MCA was an analysis, not a theory. However, I think the point has
> been well made in these discussions that MCA/MCT has axioms that lead of
> necessity to certain conclusions, and that any specific application comes
> from a conjunction of this set of axioms (and their derived consequences)
> and the set of assumptions about (or initial conditions of) a specific
> system. In this sense, it really isn't different from other things that
> are called theories, whether special relativity or the origin of species
by
> natural selection.
>> What would be the advantage of deciding to call it a theory? It would
> focus our minds on a pithy and accurate statement of what the theory is
> about. If we could agree on that, and promote it, it would be more likely
> that we could get it referred to in textbooks. I have no doubt that there
> has been a rival theory of metabolic control - it is the principle of
> control of a pathway by a single rate limiting step, and its corollaries
> that the rate limiting step is a feedback-inhibited enzyme at the
beginning
> of a pathway. MCA cannot just be the theory that says that the theory of
> the necessary existence of rate-limiting steps is wrong.
>> What could be possible contenders for the statements of MCT? I suggest:
> 1. The total amount of control of flux exertable by the enzymes of a
> metabolic pathway is limited. (No need to tie it to the particular value
of 1)
> 2. The control of metabolite concentrations is not so limited.
> 3. The control is potentially distributed over the components of the
> system in a way that depends on all the components of the system and
varies
> with the operating conditions of the pathway.
>> Corollary of 1: large changes of flux are not achieved by control at a
> single enzyme site.
> Corollary of 2 : specific control mechanisms are needed to limit
> concentration changes when fluxes change.
>> I'll be interested in reactions,
>> David
