On 03/20/2013 12:09 PM, Laura M. Tully wrote:

Sorry -  meant to respond to list.

Thanks for the response Doug - I have a few follow up questions I hope you can clarify:

      * is the xxx.ocn.annot file created by mri_glmfit-sim considered
        a "label" by freesurfer?

It is an annotation

      * or does it need to be converted to a label before I can use
        mri_segstats to extract the betas from the cluster(s) in the
        xxx.ocn.annot?

      * I tried using the following command line to attempt to extract
        the betas from the cluster annotation file created by
        mri_glmfit-sim, but received an error:

    mri_segstats --i beta.mgh --annot fsaverage hemi
    contrast/xxx.ocn.annot --avgwf avgwf.dat

That looks correct. Did it work? Add --excludeid 0 to keep it from reporting on non-clusters

      * how does one convert an xxx.ocn.annot file to a label?

If needed, use mri_annotation2label

      * If I have run 3 different glms looking at the relationship

        between thickness and 3 different behavioral variables, and
        found 1 (or more) cluster in each glm using mri_glmfit-sim can
        I create 1 annotation/label with all the clusters from the
        separate glms and then extract the betas from that one
        annotation using mri_segstats? Or do I need to treat each glm
        separately?

It is probably easiest if you treat each glm separately. You can break the annotations into separate labels, then recombine the labels into another annotation (mris_label2annot), but it is a lot of work.

     *


      * I saw on the listserv some references to matlab functions that

        can calculate r square and partial correlations for glms with
        more than one predictor variable (e.g pcc between thickness
        and behavioral variable 1; pcc between thickness and
        behavioral variable2; Rsq for overall model) but I could not
        find the .m scripts for the functions (MRIread
        or fast_glm_pcc?) - would these functions be appropriate and
        if so where might I find them?

Yes. They are in $FREESURFER_HOME/matlab and $FREESURFER_HOME/fsfast/toolbox
doug

     *

    Thanks in advance for your help!

    Laura.


    On Wed, Mar 20, 2013 at 8:29 AM, Douglas N Greve

    <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote:


        On 03/20/2013 10:17 AM, Laura M. Tully wrote:
        > Hi,
        >
        > I wanted to re-post my questions from a couple of days ago
        below, but
        > with some more specific questions following a search through the
        > archives.
        >
        > I want to be able to extract the beta values from a cluster
        identified
        > using mc-z in a group by behavioral variable interaction so
        that I can
        > 1) plot the relationship of thickness to behavioral variable
        data by
        > group in that cluster, 2) conduct post hoc tests to examine the
        > interaction, and 3) calculate the Rsquare and partial
        correlations for
        > each variable in the glm (i.e. how much variation in
        thickness is
        > explained by my behavioral variable).
        >
        > To extract the beta values from a cluster identified by mc-z
        would I
        > treat the cluster like a label and use mri_segstats to
        extract the
        > beta weights from the cluster? Would I need to make a label
        of all the
        > clusters that I want to do this for first?
        You can do it label by label. Or, if the annotation created by
        mri_glmfit-sim has all the clusters you want, you can use that
        (also in
        mri_segstats). You can also create an annotation from
        individual labels
        with mris_label2annot.
        >
        > Is there a way to calculate the basic statistics for the glm and
        > extract in table form? i.e. Fs and ps for peaks of each
        cluster? What
        > about Rsquare, or the correlation between thickness and my
        behavioral
        > variable in the clusters? or would I need to compute these
        outside of
        > freesurfer using the extracted betas?
        You will need to do it outside of FS. What are Fs and ps?
        doug

        >
        > Thank you!
        >
        > Laura.
        >
        >
        >
        >
        >
        >
        > On Mon, Mar 18, 2013 at 5:05 PM, Laura M. Tully
        > <tully.laura@googlemail.com
        <mailto:tully.laura@googlemail.com>

        <mailto:tully.laura@googlemail.com

        <mailto:tully.laura@googlemail.com>>> wrote:
        >
        >     Hi Freesurfer experts,
        >
        >     I'm hoping you can help me understand how to interpret
        >     interactions in clusters identified in whole brain
        analysis using
        >     glmfit and glmfit-sim. Below I describe what I've done
        and what
        >     I'd like to be able to do. Any suggestions would be most
        appreciated!
        >
        >       * I have two groups (patients, controls) and a behavioral
        >         variable of interest (social functioning). I am
        interested in
        >         cortical thickness differences between groups (main
        effect of
        >         group), whether cortical thickness relate to social
        >         functioning across the group (main effect of social
        >         functioning), and whether this relationship differs
        by group
        >         (group x social functioning interaction).
        >       * I ran whole brain analysis using mri_glmfit with
        group and
        >         functioning as variables of interest whilst controlling
        >         for/regressing out gender, age, and mean thickness.
        i.e. 4
        >         classes (conmale,confemale, ptmale, ptfemale) and 3
        continuous
        >         variables (age, AvgThickness, Functioning) = 16
        regressors.
        >       * I tested the group x functioning interaction with the
        >         following contrast - is it correct?
        >
        >     0000000000000.50.5-0.5-0.5
        >
        >       * I then ran mri_glmfit-sim to identify clusters that
        survive
        >         multiple comparisons. This revealed 4 clusters (3 in
        LH; 1 in
        >         RH) that represent regions showing significant group x
        >         functioning interaction.
        >       * I visualized the clusters in tksurfer, and by
        loading the
        >         y.fsgd file was able to visualize the plotted data
        to get a
        >         sense of the interaction, but this is as much as I
        know in
        >         terms of how to examine interactions in the cluster
        data......
        >
        >     My specifc questions include:
        >
        >       * I understand that the values in xxx.sig.cluster.mgh
        overlay
        >         reflect log10 p values, the signs of which indicate the
        >         direction of the relationship (i.e. -3 = negative
        correlation
        >         between thickness & variable) but I'm not sure how to
        >         interpret this in the context of an interaction with
        group?
        >       * I understand that the values in xxx.y.ocn.dat
        contain the
        >         average thickness value for each subject in that
        cluster and
        >         that in a simple between groups test this data could
        be used
        >         to conduct post hoc t-tests to show the direction of the
        >         difference, but again I'm not sure how to use this
        data in the
        >         context of the interaction. What do the values
        represent in a
        >         group x variable interaction?
        >
        >     Ideally, I'd like to extract the contrast estimates for each
        >     subject in the group x functioning contrast and plot it
        in another
        >     program and conduct pairwise comparisons (t-tests) in
        order to get
        >     a better understanding of the interaction). I'm not sure
        how to do
        >     this - is it possible? My thinking is that I do
        something similar
        >     in fMRI analysis in spm where I can plot the contrasts in a
        >     significant cluster and then extract both the average
        contrast
        >     estimates for each group and the contrast estimates for each
        >     individual subject.
        >
        >     Thanks in advance!
        >
        >     Laura.
        >
        >
        >     --
        >     --
        >     Laura M. Tully, MA
        >     Social Neuroscience & Psychopathology, Harvard University
        >     Center for the Assessment and Prevention of Prodromal
        States, UCLA
        >     Semel Institute of Neuroscience
        > ltully@mednet.ucla.edu <mailto:ltully@mednet.ucla.edu>

        <mailto:ltully@mednet.ucla.edu <mailto:ltully@mednet.ucla.edu>>
        > ltully@fas.harvard.edu <mailto:ltully@fas.harvard.edu>
        <mailto:ltully@fas.harvard.edu <mailto:ltully@fas.harvard.edu>>
        > 310-267-0170 <tel:310-267-0170> <tel:310-267-0170

        <tel:310-267-0170>>
        >     --
        >     My musings as a young clinical scientist:
        > http://theclinicalbrain.blogspot.com/
        >     Follow me on Twitter: @tully_laura
        >
        >
        >
        >
        > --
        > --
        > Laura M. Tully, MA
        > Social Neuroscience & Psychopathology, Harvard University
        > Center for the Assessment and Prevention of Prodromal
        States, UCLA
        > Semel Institute of Neuroscience
        > ltully@mednet.ucla.edu <mailto:ltully@mednet.ucla.edu>

        <mailto:ltully@mednet.ucla.edu <mailto:ltully@mednet.ucla.edu>>
        > ltully@fas.harvard.edu <mailto:ltully@fas.harvard.edu>
        <mailto:ltully@fas.harvard.edu <mailto:ltully@fas.harvard.edu>>

        > 310-267-0170 <tel:310-267-0170>
        > --
        > My musings as a young clinical scientist:
        > http://theclinicalbrain.blogspot.com/
        > Follow me on Twitter: @tully_laura
        >
        >

        > _______________________________________________
        > Freesurfer mailing list
        > Freesurfer@nmr.mgh.harvard.edu

        <mailto:Freesurfer@nmr.mgh.harvard.edu>

        > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

        --
        Douglas N. Greve, Ph.D.
        MGH-NMR Center

        greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>
        Phone Number: 617-724-2358 <tel:617-724-2358>
        Fax: 617-726-7422 <tel:617-726-7422>

        Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
        <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>
        FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
        <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>

        Outgoing:
        ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/

        _______________________________________________
        Freesurfer mailing list
        Freesurfer@nmr.mgh.harvard.edu

        <mailto: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.




    --     --
    Laura M. Tully, MA
    Social Neuroscience & Psychopathology, Harvard University
    Center for the Assessment and Prevention of Prodromal States, UCLA
    Semel Institute of Neuroscience

    ltully@mednet.ucla.edu <mailto:ltully@mednet.ucla.edu>
    ltully@fas.harvard.edu <mailto:ltully@fas.harvard.edu>
    310-267-0170 <tel:310-267-0170>
    --     My musings as a young clinical scientist:
    http://theclinicalbrain.blogspot.com/
    Follow me on Twitter: @tully_laura




--
--
Laura M. Tully, MA
Social Neuroscience & Psychopathology, Harvard University
Center for the Assessment and Prevention of Prodromal States, UCLA Semel Institute of Neuroscience
ltully@mednet.ucla.edu <mailto:ltully@mednet.ucla.edu>
ltully@fas.harvard.edu <mailto:ltully@fas.harvard.edu>
310-267-0170
--
My musings as a young clinical scientist: http://theclinicalbrain.blogspot.com/
Follow me on Twitter: @tully_laura

--
Douglas N. Greve, Ph.D.
MGH-NMR Center
greve@nmr.mgh.harvard.edu
Phone Number: 617-724-2358
Fax: 617-726-7422

Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/