How to fitting neuron output?

Using the Multiple Run Fitter, praxis, etc..
Post Reply
chalspark
Posts: 17
Joined: Tue Jun 01, 2010 1:19 am

How to fitting neuron output?

Post by chalspark » Fri Jul 08, 2011 12:28 pm

Hello.

I would like to know about how to fit neuron output through biology experiment result.

Actually, I want to made slow depolarization and delayed spiking neuron model. But it isn't easy to me.

I used many neuron model in modelDB. but, I couldn't success.

So, I would like to know that is it possible fitting neuron model results to biology experiment results?

If you know about slow depolarization and delayed spiking neuron model, please let me know.

or Could you recommend some helpful papers about this problem?

Thanks.

ted
Site Admin
Posts: 5624
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: How to fitting neuron output?

Post by ted » Fri Jul 08, 2011 11:31 pm

NEURON has a built-in tool for optimizing models (tuning parameters so that simulation results will approximate experimental results closely). It's called the Multiple Run Fitter. Look on the Documentation page http://www.neuron.yale.edu/neuron/docs for a tutorial about how to use this tool to optimize models.

chalspark
Posts: 17
Joined: Tue Jun 01, 2010 1:19 am

Re: How to fitting neuron output?

Post by chalspark » Sat Jul 09, 2011 1:09 am

Thanks for your answer.

I don't have biology experiment result. Just i have graph result in the paper.

Multiple Run fitter must need experiment result data.

Is it possible to make neuron model with graph result in paper?

How could i transfer graph result to file?

I am novice user. If my question was stupid, generously forgive me.

Thanks.

ted
Site Admin
Posts: 5624
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: How to fitting neuron output?

Post by ted » Sat Jul 09, 2011 1:22 pm

chalspark wrote:I don't have biology experiment result. Just i have graph result in the paper.
Multiple Run fitter must need experiment result data.
Is it possible to make neuron model with graph result in paper?
How could i transfer graph result to file?
You'll have to measure points from the graph yourself, or ask the authors of the paper if they could send you the data that they used to generate the graph.

chalspark
Posts: 17
Joined: Tue Jun 01, 2010 1:19 am

Re: How to fitting neuron output?

Post by chalspark » Tue Jul 12, 2011 8:17 pm

I used several channels in modelDB to optimize simulation result.

Channel conductance was changed for result. So, conductance values not equal to original model and also channels were mixed by few models.

Finally, I get the similar result. but I have a question.

Are these conductances reliable? Are these channels combination reliable? and can i used my model in the research?

ted
Site Admin
Posts: 5624
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: How to fitting neuron output?

Post by ted » Wed Jul 13, 2011 12:21 am

Important questions. They can only be answered by someone who has specific knowledge about what you are trying to do, has taken the time to examine the models in ModelDB that you used, and then made a judgement as to whether what you did is reasonable or not. Sounds like that person is you, or a close collaborator of yours. Ultimately, you'll have to explain what you did, and why, to someone who reviews your paper(s) for publication.

Post Reply