Hi Ajay,
Make sure you have to run palm_hemisplit on the outputs first, i.e.:
palm_hemisplit bh.thickness_*
Note that the "-o bh.thickness" was a somewhat poor choice of outputs as
now it's necessary to use the underscore ("_") to avoid overwriting
original inputs; for future runs, consider something as "-o bh.results" or
similar.
Then load these split files on top of the surfaces.
The different results shown using white or pial make no sense I'm afraid.
The surfaces don't contain data, just the geometry on top of which the
results are displayed. Using a different surface on Freeview doesn't change
the results.
All the best,
Anderson
On 19 September 2016 at 17:38, Ajay Kurani <dr.ajay.kur...@gmail.com> wrote:
> Hi Anderson,
> My full design contrasts are below:
>
> /ContrastName1 HC > Grp1
> /ContrastName2 HC < Grp1
> /ContrastName3 HC > Grp2
> /ContrastName4 HC < Grp2
> /ContrastName5 Grp1 > Grp2
> /ContrastName6 Grp1 < Grp2
> /ContrastName7 M > F
> /ContrastName8 M < F
> /ContrastName9 HC/Grp1 M/F Interaction
> /ContrastName10 HC/Grp2 M/F Interaction
> /ContrastName11 Grp1/Grp2 M/F Interaction
> /NumWaves 9
> /NumPoints 11
> /Matrix
> 1 1 -1 -1 0 0 0 0 0
> -1 -1 1 1 0 0 0 0 0
> 1 1 0 0 -1 -1 0 0 0
> -1 -1 0 0 1 1 0 0 0
> 0 0 1 1 -1 -1 0 0 0
> 0 0 -1 -1 1 1 0 0 0
> 1 -1 1 -1 1 -1 0 0 0
> -1 1 -1 1 -1 1 0 0 0
> 1 -1 -1 1 0 0 0 0 0
> 1 -1 0 0 -1 1 0 0 0
> 0 0 1 -1 -1 1 0 0 0
>
> My colums correspond to the following:
> EV1:HC-M
> EV2:HC-F
> EV3:Grp1-M
> EV4:Grp1-F
> EV5:Grp2-M
> EV6:Grp2-F
> EV7:Age
> EV8:Education
> EV9:Disease Severity
>
> In the folder I was running the analysis I put the lh.thickness.10mm.mgz,
> rh.thickness.10mm.mgz, lh.white (fsaverage), rh.white (fsaverage),
> lh.mask.mgh (taken from running qdec initially), rh.mask.mgh (taken from
> running qdec initially).
>
> I initially ran palm_hemimerge lh* within matlab
> Then from a terminal I ran the following command:
>
> palm -i bh.thickness.10mm.mgz -d design.mat -t design.con -o bh.thickness
> -n 500 -approx tail -corrcon -s bh.white -T -tfce2D -logp -m bh.mask.mgz
> -nouncorrected
>
> The command ran fully. When I loaded contrast 6 I found no results on the
> pial surface, however the white matter surface (where the mask was) was
> speckled all over with no cluster.
>
> If there is a location,I can upload the stats file if that is easier.
>
> Thanks,
> Ajay
>
> On Sat, Sep 17, 2016 at 10:03 PM, Ajay Kurani <dr.ajay.kur...@gmail.com>
> wrote:
>
>> Hi Anderson,
>> Thanks for the help. When viewing my results they looked very
>> strange. Upon further investigation it looks as though the mask I supplied
>> to PALM was a white matter mask (mask.mgh from running qdec initially)
>> created when I ran qdec. I assumed this would be the whole cortex but I
>> was wrong. Therefore it seems to only run permutation testing on the
>> surface of the white matter. Due to the fact that it is unsmoothed white
>> matter, I think this is why we see some speckling bleeding through near the
>> boundaries
>>
>> In order to do permutation testing accurately for surface based cortical
>> thickness, would the mask need to be a volume file which is between the
>> pial and white matter surfaces or would it just need to be the pial surface
>> (lh.pial / rh.pial), or something else? Any suggestions on the best way to
>> create this?
>>
>> Thanks,
>> Ajay
>>
>> On Sat, Sep 17, 2016 at 1:03 PM, Ajay Kurani <dr.ajay.kur...@gmail.com>
>> wrote:
>>
>>> Hi Anderson,
>>> Thanks for the help. When viewing my results they looked very
>>> strange. Upon further investigation it looks as though the mask I supplied
>>> to PALM was a white matter mask (mask.mgh from running qdec initially)
>>> created when I ran qdec. I assumed this would be the whole cortex but I
>>> was wrong. Therefore it seems to only run permutation testing on the
>>> surface of the white matter as seen in the attached photo. Due to the fact
>>> that it is unsmoothed white matter, I think this is why we see some
>>> speckling bleeding through near the boundaries
>>>
>>> In order to do permutation testing accurately for surface based cortical
>>> thickness, would the mask need to be a volume file which is between the
>>> pial and white matter surfaces or would it just need to be the pial surface
>>> (lh.pial / rh.pial), or something else? Any suggestions on the best way to
>>> create this?
>>>
>>> Thanks,
>>> Ajay
>>>
>>>
>>>
>>>
>>> On Thu, Sep 15, 2016 at 1:46 PM, Ajay Kurani <dr.ajay.kur...@gmail.com>
>>> wrote:
>>>
>>>> Hello Freesurfer Experts,
>>>> I was running permutation simulations on cortical thickness data and
>>>> I had an issue with non-orthogonal covariates with mri_glmfit-sim -perm. I
>>>> then tried FSL's PALM which is an extension of randomize to calculate
>>>> threshold free stats. I saved the output as logp(which is similar to qdec
>>>> I believe), however I have not been able to load the stats files
>>>> correctly. The output of palm is lh.thickness_tfce.mgz for my various
>>>> contrasts.
>>>>
>>>> 1) Is .mgz the proper format for the stats files or do I need to
>>>> convert this to another type like .mgh etc?
>>>>
>>>> 2) Can I display this in freeview or is another program needed? I also
>>>> tried tksurfer but when I loaded the stats file as an overlay nothing
>>>> displayed. I want to make sure that the stats is loaded as an overlay in
>>>> freeview/tksurfer and if so, do I need to select anything special so that
>>>> it scales the logp values correctly?
>>>>
>>>> Thanks,
>>>> Ajay
>>>>
>>>
>>>
>>
>
> _______________________________________________
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>
>
> 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.
>
>
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
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.
