Hi, friends
i want to do m'=am(1-m)-bm*m as m'=am(1-m)-bm*m+gp
taht is to say
from m'=am(1-m)-bm*m to m'=am(1-m)-bm*m+gp
how change m'=am*(1-m)-bm*m
-
okanerkaymaz
it's OK
Unfurtunately i writen shuc that code that is folawing
/* this is part of specification of biological properties */
Area = 100 // um2
Nasingle=0
Ksingle=0
objref u,f
func gauss(){
m=0 s=1
u = new Random()
geu=u.uniform(0,1)
z=s*sqrt(2*log(1/(1-geu)))
feu=u.repick()
x=m+z*sin(2*PI*feu)
return x
}
// argument should be desired area in um2
func varyansm() {
varm=(2/Nasingle)*(Nahh[0].m.am*Nahh[0].bm)/(Nahh[0].m.am+Nahh[0].m.bm)
return varm
}
func varyansh() {
varh=(2/Nasingle)*(Nahh[0].ah*Nahh[0].bh)/(Nahh[0].ah+Nahh[0].bh)
return varh
}
func varyansn() {
varn=(2/Ksingle)*(Khh[0].an*Khh[0].bn)/(Khh[0].an+Khh[0].bn)
return varn
}
vma=varyansm()
vh=varyansh()
vn=varyansn()
proc noise(){
gu=gauss()
gm=sqrt(vma)*gu
Khh[0].m=Khh[0].m+gm //i don't know what i will write here
gh=sqrt(vh)*gu
Nahh[0].h=Nahh[0].h+gh //i don't know what i will write here
gn=sqrt(vn)*gu
Khh[0].n=Khh[0].n+gn//i don't know what i will write here
}
also i run this hoc file but it doesn't behave like stochastic HH
you think what have i to do about this event?
Unfurtunately i writen shuc that code that is folawing
/* this is part of specification of biological properties */
Area = 100 // um2
Nasingle=0
Ksingle=0
objref u,f
func gauss(){
m=0 s=1
u = new Random()
geu=u.uniform(0,1)
z=s*sqrt(2*log(1/(1-geu)))
feu=u.repick()
x=m+z*sin(2*PI*feu)
return x
}
// argument should be desired area in um2
func varyansm() {
varm=(2/Nasingle)*(Nahh[0].m.am*Nahh[0].bm)/(Nahh[0].m.am+Nahh[0].m.bm)
return varm
}
func varyansh() {
varh=(2/Nasingle)*(Nahh[0].ah*Nahh[0].bh)/(Nahh[0].ah+Nahh[0].bh)
return varh
}
func varyansn() {
varn=(2/Ksingle)*(Khh[0].an*Khh[0].bn)/(Khh[0].an+Khh[0].bn)
return varn
}
vma=varyansm()
vh=varyansh()
vn=varyansn()
proc noise(){
gu=gauss()
gm=sqrt(vma)*gu
Khh[0].m=Khh[0].m+gm //i don't know what i will write here
gh=sqrt(vh)*gu
Nahh[0].h=Nahh[0].h+gh //i don't know what i will write here
gn=sqrt(vn)*gu
Khh[0].n=Khh[0].n+gn//i don't know what i will write here
}
also i run this hoc file but it doesn't behave like stochastic HH
you think what have i to do about this event?
I don't know the answer to your question, so just a few search tips. If you implement your mechanism in mod, which I think is what you are doing, you should try to read up on NMODL and read up on the numerical integration schemes used. When you did that you should try to decide which integration scheme is suitable and whether you can set gp in the mod-file.
Anyway if you draw a new value of gp from the gaussian distribution at each timestep it seems to me that only first order integration schemes can be applied.
Anyway if you draw a new value of gp from the gaussian distribution at each timestep it seems to me that only first order integration schemes can be applied.
Our postings crossed, which gives an odd appearance now. If you implement your algorithm in hoc you have to make sure that your stochastic variables gets drawn at every timestep. This can be done by plugging in to the standard run system, please consult The NEURON Book by Carnevale and Hines on this topic. It seems to me highly unlikely that this will yield a very fast simulation, which seems to be part of the purpose of the Fox-Lu algorithm.
So I think you will have to move over to NMODL and solve the problem there.
So I think you will have to move over to NMODL and solve the problem there.