IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

The importance of detaile mechanism

Herbert M Sauro hsauro at cds.calxxxtech.edu
Tue Aug 22 18:41:39 EST 2000



Mechanism and MCA/MCT

As far as I am aware I think the reason why 'mechanism' (which I interpret
to mean detailed enzyme mechanism) is apparently reduced in importance in
MCA/MCT is related to the question, in what terms should we describe system
properties, and at what detailed level of description do we need in order to
describe higher-level behaviour. Many of the things one observes in whole
metabolic systems do not require detailed descriptions of enzyme mechanisms,
in fact all one needs is an input/output response for each enzyme, the
elasticities in MCA terminology. I think there is little to be gained by
examining in the detail the mechanism of an enzyme, this does not mean that
one cannot, it just depends on your question.

In designing an electronic circuit, one only needs to know the input/output
characteristics of the individual components in order to 'understand' the
origins of the complete circuit behaviour. In terms of an oscillating
metabolic system one will not find a source of oscillation but rather a set
of properties of enzymes which when taken together result in the emergence
of oscillation. There is probably sufficient information in the set of
elasticities for each enzyme in the pathway to workout whether it might
oscillate or not (by forming the jacobian etc) and even to understand why it
should oscillate. Perhaps the question you're asking is at another level,
namely explain why a particular enzyme has the set of elasticity values it
has, to answer this question one *would* need to consider the detailed
enzyme mechanism.

Thinking in terms of large many interaction systems is traditionally very
difficult in any discipline, thinking is terms of the individual parts of
the system (individual enzymes) is perhaps 'easier'. I suspect that the
emphasis on mechanism in traditional biochem/mol biol is partly because of
this. There does need to be a change in the way biochemists and molecular
biologists think about cells, they need to think not in terms of individual
enzymes, genes etc but in terms of collective behaviour and to do this it is
best (IMHO) to start the level of description at a level higher than raw
enzyme mechanism, one such possible level is at the level of elasticities.
Interestingly the connectivity theorem which relates system to local
properties contains no suggestion of enzyme mechanism within it.

Iin fact, two pathways with exactly the same elasticity profiles but
different ground level enzyme mechanisms should in theory behave
*identically*, in this sense the mechanism is of no importance.

The above was probably a rather long winded way of saying something which is
probably quite simple but I hope it makes sense!

As for why should MCA be useful, I guess it depends on your questions. My
main question has always been, why do pathways behave the way they do, and
can this behaviour be explained in terms of the constituent parts of the
system. For me, MCA's greatest contributions lie in two areas:

1. It helps to focus my mind and organise what appears to be a hopelessly
complex problem
2. It allows me to describe system behaviour in terms of components
properties

There may well be, and indeed I hope there will be, other and more novel
approaches to understanding metabolic systems but I ain't seen any yet.

You're right in saying that Garfinkel's modelling used detailed mechanisms
in their simulations. I think somewhere I have a copy of BIOSIM and I
remember seeing that this is what they did. For some analyses this level of
analysis is not necessary, it is possible that metabolic transitions may be
time delayed or enhanced depending on the mechanisms, but the final steady
state shouldn't be influenced by which model (detailed or black-box rate)
one chooses.

Herbert Sauro

"Petr Kuzmic" <pkuzmic at biokin.com> wrote in message
news:39A27B9E.EBEAE4E3 at biokin.com...
>
>
> Herbert M Sauro wrote:
> [...]
> > One thing his [David Garfinkel's]
> > efforts taught me was that simulation in a vacum of formal theory is
almost
> > useless, unless perhaps you're an engineer in which case you might only
be
> > interested in input and outputs. But as a scientist one needs to
understand
> > what's going on inside and for that one needs some sort of organising
> > framework, and for me that turned out to be MCA/MCT.
> [...]
>
> I am always trying to understand more about the important work that's
> been done by people belonging to the Metabolic Control Theory/Analysis
> school of thought.  After having read Herbert's statement above, I am
> now wondering exactly in what sense does MCA contribute to a greater
> "understand[ing of] what's going on inside".   I'd like to contrast what
> Herbert wrote with the following statement, taken from Athel's FAQ page:
>
> "Metabolic control analysis tends to treat the kinetic properties of the
> component enzymes as a black box. Some authors have been very critical
> of this, suggesting that shedding light on mechanism is the only reason
> for studying kinetics in the first place. [...] Studies of biochemical
> kinetics have been dominated for nearly a century by an interest in
> molecular mechanisms, but for understanding how whole pathways behave it
> has been found useful to decrease the emphasis on mechanism. [...]"
>
> Am I missing something?  Isn't this "black box" approach focusing at
> "inputs and outputs" and not paying too much attention to "what's going
> on inside"?  I believe that every step in Garfinkel's simulations was
> formulated in terms of the underlying *mechanism*.
>
> -- Petr
>
> _____________________________________________________________________
> P e t r   K u z m i c,  Ph.D.               mailto:pkuzmic at biokin.com
> BioKin Ltd. * Software and Consulting           http://www.biokin.com







More information about the Btk-mca mailing list

Send comments to us at biosci-help [At] net.bio.net