Page 1 of 1

### Channel noise

Posted: Sun Aug 26, 2018 4:50 am
Hi guys,

One knows, an ionic channel can be either open or closed - if we take a look at N channels, the probability that n channels are open needs to follow a binomial distribution.... now the standard-deviation is proportional to the square root of the considered channels -- therefore : more channels are producing noise proportional to the square root .... I read (in some papers) it should be proportional to the inverse of the square-root... but I don't see the logic behind - could you please give me a short explanation or do you know papers explaining this certain issue ?

Thank you in advance and kind regards

### Re: Channel noise

Posted: Sun Aug 26, 2018 4:21 pm
Time to learn about statistics and stochastic processes.

### Re: Channel noise

Posted: Sat Sep 29, 2018 3:53 am
Dear Ted,

I know an stochastic approach with Markov chains does exist, but is the question answerable due to simple statistics ?

If we take a look at N channels - the probability that n channels are open follows a binomial distribution Bin(n,p) - so the standard deviation is proportional to the square root of n.
If we take a look at the relative frequency it is proportional to the reciprocal value.

Is this the answer in a very intuitive way ?

Kind regards,

### Re: Channel noise

Posted: Sun Sep 30, 2018 12:57 pm
That's it.

### Re: Channel noise

Posted: Sat Mar 09, 2019 12:08 pm
Hi,

I would like to propose an other approach:

lets assume we divide a membrane into n parts, which are represented by random variables X_1,...,X_n - the "value" of this random variables is representing the current fluctuations (without specifying the distribution, but maybe we can assume a N(0,1) distribution.)

When summing the n parts, the standard deviation of

\sum_{I=1}^{n}X_i should be growing proportional to the square root of n.

I would like to describe the standard deviation of this sum under the condition that the total membrane potential V(t) is influencing the fluctuations of the X_i.

Do you know how to model this ?

kind regards