What is an FCC (and logs)?

hrs at ABERYSTWYTH.AC.UK hrs at ABERYSTWYTH.AC.UK
Sun Jun 20 06:30:52 EST 1993


| >Hmmm... I wonder if taking the log (base 2 of course) of the FCC's would
| >suggest ...
| 
| >  Tom Schneider
| 
| Probably nothing, because the FCCs are already in log form (to the base 2
| of course). The FCCs are ratios of fractional changes of a variable
| with respect to a parameter and I shouldn't think one could draw much
| more info by logging again.

>So - not knowing about FCC's - there is the possibility that these Flux Control
>Coefficients can _already_ be interpreted as having units of bits!  Two
>questions come up: what exactly is a FCC (besides Frederick Community College)
>and whether one can interpret them as measuring some kind of choice being made
>by the system.
>
>  Tom Schneider

The log to the base 2 was just a joke I'm afraid, perhaps I've raised
your hopes when I shouldn't have. The FCCs are very simple, they are just
sensitivity measures. They measure how sensitive a system variable is
to changes in a system parameter. For a metabolic pathway, a common
variable to consider is the flux through the pathway, and a common
parameter is often the concentration of a particular enzyme, if we were
to extend the system out to the genome, then of course the enzyme
concentration would no longer be a parameter but a variable and the
parameter would now be something like promoter efficiency or something.
A log (to the base e) is implied in their definition because instead
of just taking, dV/dP we actually consider the scaled form (dV/V)/(dP/P)
which can be interpreted as dlnV/dlnP. Scaling has a number of advantages,
an especially good one for experimentalists is that the units of
measurement disappear, so one could measure flux in units of 'movement
in needle across paper' instead of having to get down to moles per sec per
etc etc.

I think the really important point made by metabolic control analysis
is that it manages to connect things like FCCs (which are whole system
properties, they can only be measured in the intact 'living'  pathway) 
to another type of coefficient (the elasticity) which is a
local enzyme property (related to basic things like Kms and so on) and
can be measured in the test tube so to speak.

This means that we can now understand the relationship between the effect
of changing the level of a gene expression on a phenotype (eg a flux) and
the  basic enzyme properties (Kms, Ki, etc), i.e relate genotype to
phenotype. This advance is of sufficient importance that I think it warrants
the creation of a new group called btk-mca.

Whether one could apply the info theory to mca is another matter though.
The main reason why I got into mca was that it gave me a way of working out
the complex relationships that exist in a metabolic pathway between the
enzymes, metabolites, signals etc. Solutions to the basic differential
equations are not possible and computer simulation left me in the dark
as much as the real thing so mca turned out to be what I was looking for.

Hope this is of help.

Herbert Sauro
Penrodyn,
Pontrhydygroes
Ystrad Meurig
Dyfed, SY25 6DP, UK



More information about the Bioforum mailing list