Hi Stephen,
The p-value images will not be useful here ( for one permutation no sensible probabilistic inference is possible ). In your original email it sounded like you only wanted the t-statistics, these will be the Stage_3_smoothed_ic0000_tstat<blah> images.
Hope this helps,
Kind Regards
Matthew
--------------------------------
Dr Matthew Webster
FMRIB Centre
John Radcliffe Hospital
University of Oxford
> On 10 Dec 2020, at 03:43, Suss, Stephen <[log in to unmask]> wrote:
>
> Sorry, here is the script and the output. As you can see it is saying the p values are zero:
>
> --
> WORKDIR=`readlink -f ${!#}`
> cd $WORKDIR
>
>
> for file in ${WORKDIR}/dr_stage2_ic*
> do
> sub=`basename $file | cut -d _ -f 4`
> randomise -i $file -o Stage_3_np_${sub} -d design.mat -t design.con -n 1 -m mask -T -R
> done
>
> --
> susssj2@icare:/data2/rsprojects/DualRegression_outputs/ICA_Early_v_Prolonged_AgeGender3_Smith09_1r.dr$ for i in Stage_3_smoothed_ic00??_tfce_corrp_tstat?.nii.gz ; do
> > echo $i `fslstats $i -R`;
> > done
> Stage_3_smoothed_ic0000_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0000_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0001_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0001_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0002_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0002_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0003_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0003_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0004_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0004_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0005_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0005_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0006_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0006_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0007_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0007_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0008_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0008_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0009_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
> Stage_3_smoothed_ic0009_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
>
> In this case I smoothed data before entry, but I am experiencing same error with unsmoothed data as well.
>
> Best,
> Steve
>
> From: Suss, Stephen <[log in to unmask]>
> Sent: Wednesday, December 9, 2020 10:37 PM
> To: [log in to unmask]
> Subject: Re: [FSL] Simple group comparison of ICA data without permutation
>
> I Actually got it working, the only issue when I run it, all the P values come back as zero. Here is the script I am running and the out put.
>
> WORKDIR=`readlink -f ${!#}`
> cd $WORKDIR
>
>
> for file in ${WORKDIR}/dr_stage2_ic*
> do
> sub=`basename $file | cut -d _ -f 4`
> randomise -i $file -o Stage_3_np_${sub} -d design.mat -t design.con -n 1 -m mask -T -R
> done
>
>
>
> Stephen (Steve) J. Suss Jr., BS
>
> UPMC Research Specialist
>
> Loeffler Building
>
> 121 Meyran Avenue
>
> Pittsburgh, PA 15260
>
> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Suss, Stephen <[log in to unmask]>
> Sent: Wednesday, December 9, 2020 9:00 AM
> To: [log in to unmask] <[log in to unmask]>
> Subject: Re: [FSL] Simple group comparison of ICA data without permutation
>
> Great! Thank you, Paul.
>
> I'm just trying to figure out the script for this case. I found the template below, but have a few questions.
>
> randomise -i TwoSamp4D -o TwoSampT -d design.mat -t design.con -m mask -T
>
> 1) From what I am reading I assume the "TwoSample4D" should be my melodic group ICA map?
> 2) Then "TwoSampT" be my stage 2 input? If so, am I to set a directory or should I use a wildcard.
> 3) What would be my mask in this case?
>
> Any help here would be much appreciated!
>
> Best,
> Steve
>
> Stephen (Steve) J. Suss Jr., BS
>
> UPMC Research Specialist
>
> Loeffler Building
>
> 121 Meyran Avenue
>
> Pittsburgh, PA 15260
>
> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of paul mccarthy<[log in to unmask]>
> Sent: Wednesday, December 9, 2020 4:13 AM
> To: [log in to unmask] <[log in to unmask]>
> Subject: Re: [FSL] Simple group comparison of ICA data without permutation
>
> Hi Steve,
>
> You can still use randomise, but just set the number of permutations to 1, and use the -R option to tell it to output the raw t-statistic images.
>
> Paul
>
> On Wed, 9 Dec 2020 at 01:05, Suss, Stephen <[log in to unmask]> wrote:
> Hi all,
>
> I apologize in advance for this basic question. We are running ICA analysis and have a relatively small data set. We would like to run a simple t-test without randomise to avoid the permutations. What would be the best way to do this? We are inputing the stage 2 data from dual regression. This is just a very basic test we are trying to run quickly.
>
> Best,
> Steve
>
> Stephen (Steve) J. Suss Jr., BS
>
> UPMC Research Specialist
>
> Loeffler Building
>
> 121 Meyran Avenue
>
> Pittsburgh, PA 15260
>
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
>
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
>
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
>
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
>
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|