permutation clustering test using TFCE

✉️ Mailing List Archive (read-only)
1

Message: 2
Date: Thu, 31 Jul 2014 15:14:01 +0200
From: Denis-Alexander Engemann <denis.engemann at gmail.com>
Subject: Re: [Mne_analysis] permutation clustering test using TFCE
To: Discussion and support forum for the users of MNE Software
        <mne_analysis at nmr.mgh.harvard.edu>
Message-ID:
        <
CA+MN3OtGzf02SVJkYNsFgwjh8-306VC7nmW8S0uc7szDq-Tr5A at mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Thanks Latitia,

>
> Hi Denis,
>
> I tried the spatio-temporal clustering with TFCE
(spatio_temporal_cluster_1samp_test) on alpha power data (size of epoch :
2s; decimation = 4 (so 501 time points)) with n_jobs = 6 and the default
TFCE parameter (dict(start = 0, step = 0.2)).

I think we need to improve the documentation on TFCE a bit. A good default
range is probably

dict(start=2, step=0.2)

>
> According to script output, 48 thresholds were used from 0 to 9.4.
> After 5h (!!), 10262484 clusters were found and finally after some others
hours the script crashed before the end ("cannot allocate memory")...
>

For TFCE N clusters equals N features. Howver if you do not scan the enitre
range of the test statistic you wont have to wait thast long.

I tried dict(start=2, step=0.2) and dict(start=2, step=0.5), but I still
had 5 120 000 clusters (15 threshold for step =0.5). How can I not scan the
entire range of the test, as you suggested ? I didn't understand what you
mean by "features"...

>
> I tried on shorter data (1s so 250 time points) but it was also too long
and too memory-demanding. Then I tried to change the step parameter to
decrease the number of thresholds to test. I took step =0.5. 17 thresholds
were used from 0 to 8. 5121000 clusters were found and my script also ended
up crashing. So... what do you suggest to get acceptable computation time?
>

See above. Btw for roughtly 10.000 clusters with 15-20 thresholds I'm
waiting for roughly 15-20 minutes per iteration (multiple iterations with
step_down=0.05).

>
> By the way, with p_threshold =0.001, I got no cluster; with p_threshold
=0.01, I got one occipito-parietal cluster (pvalue = 0.026) lasting around
500ms, after something about 10min of computation.
>

I'm not sure what you refer to by `p_threshold`. Either you pass a dict or
a float. The latter will be a classical cluster permuation analysis, the
former TFCE.
You can howver use the p_threhold as start value for TFCE.

Yes I was talking about classical permutation cluster analysis.

>
> Hoping that it is useful for you,
> Best,
> Laetitia G.

Yes, thanks!
Best,
Denis

Thanks a lot for your advice!
Laetitia
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20140804/2478822f/attachment.html

2

Hi Laetitia

Message: 2

Date: Thu, 31 Jul 2014 15:14:01 +0200
From: Denis-Alexander Engemann <denis.engemann at gmail.com>
Subject: Re: [Mne_analysis] permutation clustering test using TFCE
To: Discussion and support forum for the users of MNE Software
        <mne_analysis at nmr.mgh.harvard.edu>
Message-ID:
        <
CA+MN3OtGzf02SVJkYNsFgwjh8-306VC7nmW8S0uc7szDq-Tr5A at mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Thanks Latitia,

>
> Hi Denis,
>
> I tried the spatio-temporal clustering with TFCE
(spatio_temporal_cluster_1samp_test) on alpha power data (size of epoch :
2s; decimation = 4 (so 501 time points)) with n_jobs = 6 and the default
TFCE parameter (dict(start = 0, step = 0.2)).

I think we need to improve the documentation on TFCE a bit. A good default
range is probably

dict(start=2, step=0.2)

>
> According to script output, 48 thresholds were used from 0 to 9.4.
> After 5h (!!), 10262484 clusters were found and finally after some
others
hours the script crashed before the end ("cannot allocate memory")...
>

For TFCE N clusters equals N features. Howver if you do not scan the
enitre
range of the test statistic you wont have to wait thast long.

I tried dict(start=2, step=0.2) and dict(start=2, step=0.5), but I still
had 5 120 000 clusters (15 threshold for step =0.5). How can I not scan the
entire range of the test, as you suggested ? I didn't understand what you
mean by "features"...

samples : subjects or trials or observations
feaures : measured value at time t, location l, condition c, frequency f,
etc.

dict(start=2, step=0.2) would not scan the entire range since you start at
a value of 2. It depends on your effect size and your test statistic. with
an f-test you might want to start at 4 (roughly the point where values form
an f-dist are considered significant). Also if the maximum of your primary
test statistic is rather high, e.g. 80, you might want to jump in steps of
.05 or even 1.

I'm currently using it like that (dict(start=4, step=0.5) in sensor space
analysis and with 17640 clusters, 7 jobs and I'm waiting about 6-7 minutes
for a result using a repated measures anova as stat function (slower than
t-test) .

> I tried on shorter data (1s so 250 time points) but it was also too long
and too memory-demanding. Then I tried to change the step parameter to
decrease the number of thresholds to test. I took step =0.5. 17 thresholds
were used from 0 to 8. 5121000 clusters were found and my script also
ended
up crashing. So... what do you suggest to get acceptable computation time?
>

See above. Btw for roughtly 10.000 clusters with 15-20 thresholds I'm
waiting for roughly 15-20 minutes per iteration (multiple iterations with
step_down=0.05).

>
> By the way, with p_threshold =0.001, I got no cluster; with p_threshold
=0.01, I got one occipito-parietal cluster (pvalue = 0.026) lasting around
500ms, after something about 10min of computation.
>

I'm not sure what you refer to by `p_threshold`. Either you pass a dict or
a float. The latter will be a classical cluster permuation analysis, the
former TFCE.
You can howver use the p_threhold as start value for TFCE.

Yes I was talking about classical permutation cluster analysis.

Ah ok, got it now.

>

> Hoping that it is useful for you,
> Best,
> Laetitia G.

Yes, thanks!
Best,
Denis

Thanks a lot for your advice!
Laetitia

_______________________________________________
Mne_analysis mailing list
Mne_analysis at nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis

The information in this e-mail is intended only for the person to whom it
is
addressed. If you believe this e-mail was sent to you in error and the
e-mail
contains patient information, please contact the Partners Compliance
HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in
error
but does not contain patient information, please contact the sender and
properly
dispose of the e-mail.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20140804/9c3f08b1/attachment.html