The max is  the -log10(p), bounded below by -log10(CWP) for positive correlations Hence with an  -log10(CWP) of 4, the Max
will be greater than 4, and will specify the maximum -log10(p) value in that cluster, occuring at VtxMax
Is this right for -log10(p) kind if input? This is how i have been interpreting my results so far, and i just want
to be sure......
I was also wondering why -log10(p) transformation has been used for this kind of visualization? SPM displays
SPM_{z} scores, and that seems a better way to report the spms. Is -log10(p) an approximation to z-scores ?
sid.

On Thu, Jun 4, 2009 at 9:53 AM, Judith Segall <jsegall@mrn.org> wrote:
Thanks, for you responses Doug and Pratap. But, both of your responses need more clarification, because I believe that we are not all referring to the same 'max'.

Doug:

  I understand that when I display the sig.cluster.mgh that the a CWP value of 0.0001 would be displayed as a 4 by the color bar in tksurfer, since it is the -log10(p). However, in this case, which is about the cluster growing summary, the -log10(p) of the CWP does not equal the Max. Max is -3.312 and the CWP is 0.08250. Or an additional example form another sig.cluster.summary, where the Max is -4.310 and the CWP is 0.0192.

ClusterNo  Max   VtxMax   Size(mm^2)  TalX   TalY   TalZ    CWP      CWPLow    CWPHi    NVtxs   Annot
   1           -4.310   40526    376.78        -12.0  -66.1    5.4    0.01920  0.01750       0.02100   706      lingual

In Pratap's email his answer to my question of what is the max in the CWP is,  "Yes thickness. The MAX indicates the maximum -log10(pvalue) in that cluster." Which, is stating that it is both the thickness and the pvalue. I really think that it is thickness measure, but need some clarification from mgh.

So, is the Max in column 2 of the cluster growing summary report, from monte carlo simulations, thickness?

Thanks,

Judith S.

On Thu, Jun 4, 2009 at 9:25 AM, Douglas N Greve <greve@nmr.mgh.harvard.edu> wrote:
It will be whatever you used as input. In this case, it is the -log10(p) value of the maximum in the cluster (not the thickness value).

doug


Pratap Kunwar wrote:
Hi Judith,

Yes thickness. The MAX indicates the maximum -log10(pvalue) in that cluster.

 
Hi all. After looking over our monte-carlo results, I a simple question
about the sig.cluster.summary output.

Is the " Max" in column 2 below a measurement of thickness?
thickness, is the correct interpretation the following: the most
significant
cluster is in the laterorbitofrontal region.
   
yes at lateral orbital frontal.
The max vertex in that
 
cluster
has an average thickness of 3.312 mm and the negative directionality
indicates that it is thinner in that region than in the fsaverage (from
freesufer) used in the group analysis/simulations?
   
yes you are correct.
 
If it does not refer to thickness, is it a statistic, etc?

# Minimum Threshold 2
# Maximum Threshold infinity
# Threshold Sign    abs
# AdjustThreshWhenOneTail 1
# Area Threshold    0 mm^2
# CSD thresh  2.000000
# CSD nreps    10000
# CSD simtype  mc-z
# CSD contrast lh-Avg-thickness-CAQ-CCI
# CSD confint  90.000000
# Overall max 2.05497 at vertex 92846
# Overall min -3.3122 at vertex 60632
# NClusters          11
# Total Cortical Surface Area 65416.6 (mm^2)
# FixMNI = 1
#
# ClusterNo  Max   VtxMax   Size(mm^2)  TalX   TalY   TalZ    CWP
CWPLow    CWPHi   NVtxs   Annot
  1       -3.312   60632    278.00    -15.2   47.7  -15.4  0.08250
0.07900  0.08600   341  lateralorbitofrontal
  2       -3.231   51202    118.54    -39.5   24.9   16.9  0.53620
0.52980  0.54260   212  rostralmiddlefrontal
  3       -3.229   67407    163.06    -19.2   12.0   46.9  0.36360
0.35740  0.36980   337  superiorfrontal
  4       -2.463  117667     37.51    -47.8  -50.8    8.0  0.63240
0.62620  0.63860    79  bankssts
  5       -2.379  151243     12.92    -30.9  -42.8   36.6  0.63240
0.62620  0.63860    38  superiorparietal

I am trying to finish writing up my results and just need answer to this
question to continue.

Thanks,

Judith S.
Mind Research Network
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
   

_______________________________________________
Freesurfer mailing list
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
Phone Number: 617-724-2358 Fax: 617-726-7422

In order to help us help you, please follow the steps in:
surfer.nmr.mgh.harvard.edu/fswiki/BugReporting




_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer