[Eeglablist] Study clustering on a specific frequency band

[Makoto Miyakoshi]

Dear Maya,

> Due to different ERSP parameters gives different clustering layouts I
cannot compare directly the networks of different frequency bands.

Then cluster ICs by one ERSP (or spectrum or dipole location) parameter,
and compare activities across different frequency bands. We usually take
that approach.

Makoto

On Thu, Aug 18, 2016 at 4:36 PM, Maya Pinacle <mayapinacle at gmail.com> wrote:

> Hi Makoto,
>
> Thank you very much for the answers. The main issue that I am having is I
> would like to compare the behaviours of the networks between the clusters
> for the different frequency bands as well. Due to different ERSP parameters
> gives different clustering layouts I cannot compare directly the networks
> of different frequency bands.
>
> If I filter the data before hand the clustering, I also doub that I might
> end up with different clustering layouts.  Hence in such a case, would you
> recommend cluster the components in a broad band ERSP and then filter the
> data to the frequency band of interest and run the connectivity analysis?
>
>
> Thank you very much in advance,
>
> Maya
>
> 2016-08-18 22:51 GMT+01:00 Makoto Miyakoshi <mmiyakoshi at ucsd.edu>:
>
>> Dear Maya,
>>
>> > So for a clustering on a specific frequency band should I filter each
>> subject's data on that specific frequency band before creating the study
>> and then to run clustering on that data?
>>
>> That works.
>>
>> > or Is there a way to run clustering on a specific frequency band
>> setting the preclustering frequency parameters to the frequency band of
>> interest?
>>
>> You can specify frequency range in ERSP when clustering. It should be
>> easier for you.
>>
>> However... does ICA identify gamma? I doubt it. This is because 1) gamma
>> has small amplitude 2) gamma does not have a broad network. When I found
>> gamma, it was almost always coupled with theta (and it could have been just
>> micro tremor of eyes...)
>>
>> Makoto
>>
>>
>>
>> On Wed, Aug 17, 2016 at 1:11 PM, Maya Pinacle <mayapinacle at gmail.com>
>> wrote:
>>
>>> Dear All,
>>>
>>>
>>> I have a Study file contain datesets from 23 subjects, and I would like
>>> to cluster the ICs in the alpha frequency band and gamma frequency band
>>> seperately. Hence this will give me the chance to compare the differences
>>> between the frequency bands' behaviours on clustering.
>>>
>>> However I am a bit confused with the way of clustering on a specific
>>> frequency band so I would really appreciate your recomendations on that. So
>>> for a clustering on a specific frequency band should I filter each
>>> subject's data on that specific frequency band before creating the study
>>> and then to run clustering on that data? or Is there a way to run
>>> clustering on a specific frequency band setting the preclustering frequency
>>> parameters to the frequency band of interest?
>>>
>>> I would really appreciate any comments on that please.
>>>
>>> Maya
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.uc
>>> sd.edu
>>> For digest mode, send an email with the subject "set digest mime" to
>>> eeglablist-request at sccn.ucsd.edu
>>>
>>
>>
>>
>> --
>> Makoto Miyakoshi
>> Swartz Center for Computational Neuroscience
>> Institute for Neural Computation, University of California San Diego
>>
>
>


-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20160819/e0f98c93/attachment.html>