KINETIC Block

  > Where can I find documentation for kinetic blocks, explaining where
  > f_flux and b_flux are defined, for example, and how << and <->
  > and ~ and (c1,c2) work. I think I understand most  of it, but especially
  > the part about b_flux and f_flux I'd like to see spelled out somewhere.
  > The implementation of a kinetic scheme for the sodium potassium pump
  > is not so straightforward as the calcium pump, as some of the rate
  > constants are voltage dependent, and the Hill coeficient of the dependence
  > of the pump current on Nai is 1.4, though three Na ions are involved in the
  > reaction, of course. Anyway , thanks for any information on documentation
  > that you can give me- can the model accomodate rate constants that are a
  > function of voltage and reactions where there is cooperativity between the
  > binding sites?

	KINETIC nakpump {
		COMPARTMENT area { pump nkpump}
		COMPARTMENT voli { nai ki }
		COMPARTMENT volo { nao ko }
		~ nai << (-(ina - last_ina_pmp)*vol2surfaceratio/FARADAY
		~ 3 nai + 4 ko + pump <-> nkpump	(f1rate(v), b1rate(v))
		~ nkpump <-> 3 nao + 4 ki		(f2rate(v), b2rate(v))
		nai_pmp = unitfactor*FARADAY*(f_flux - b_flux)/area
	}

----------
Subject: Re:  COMPARTMENT statement in SCoP

COMPARTMENT statement syntax is:

        COMPARTMENT size {scalar state list}   or
        COMPARTMENT index, size(index) {vector state list}
This allows the user to select compartment sizes for states such that
quantity of matter is conserved but state units are still convenient.
The invariant that is important is
        stateunits * sizeunits / time (or indepunits)
==      fluxunits
==      rateunits * (product of stateunits^stoichiometry)
        
It is important that all reactions have the same units or that you
know exactly what you are doing.

Some examples of the use of COMPARTMENT are when:
        diffusion in cylinder with variable diameter or dx
        transfer of matter between compartments of differetnt size
        reaction between substance on a surface and substance in solution
 
        It is convenient that states have different units.

There is one caveat involving the relationship beteen COMPARTMENT and CONSERVE
Compartment sizes are taken into account on the state side of the CONSERVE
statement, but the user must explicitly specify the ammount conserved.
For example, if state A is in compartment of size a, and B is relative to
compartment of size b, and the total substance of A and B is conserved
with total = a*A + b*B then
        CONSERVE A + B = a*A0 + b*B0
is the proper statement

It is very important that you check the units with modlunit to make sure
that the programs idea of what you are doing is consistent with yours.

-----------------
Subject: Re:  COMPARTMENT directive
        
The COMPARTMENT directive  help. It is just that the compartment
units are dimensionless.  The definition of a kinetic equation is that
A_comp1 * k1 is the flux leaving A_comp1 and entering A_comp2 due to this
equation.  The change in A_comp2 per unit time is
compartment_A_comp2 * A_comp2' = sum of all fluxes entering - sum of
all fluxes leaving.

This can make the units of the k's fairly cumbersome since they depend on
the units of the compartment size and the units of the states.
i.e. all fluxes are material (extensive) quantities.
The alternative was to keep the k's in intensive units and
scale the fluxes by the appropriate compartment sizes but that was
deemed less efficient.

--------------

The COMPARTMENT syntax has two cases. One for scalar states and one for
vector states.
COMPARTMENT index, volume(index) { list of vector states without
brackets }
COMPARTMENT volume { list of scalar states }

If you understand bnf syntax you can look in nrn/nmodl/SRC/parse1.y
to see how it is parsed. In fact that file is the definitive place for
all nmodl syntax. Look at the lines around
        compart:

What it means is that states are located in different size compartments and
the that has to be taken into account when calculating the change in a state
per unit time.
Thus

COMPARTMENT v1 {a}
COMPARTMENT v2 {b}
~ a<->b (k1,k2)

solves the equations
v1*a' = -k1*a + k2*b
v2*b' = +k1*a + k2*b

Without the COMPARTMENT statements the equations are
a' = -k1*a + k2*b
so compartments affect the rate units since with compartments the  
fluxes are in extensive units (eg, moles/sec)



-----------

KINETIC 
{
  []
  ~   [ + ...] <-> ... (kf, kr)
  []
}