18 Mar
2009
18 Mar
'09
6:53 p.m.
Devdutta W wrote:
Hi Doug, This is a follow up question regarding display issue from mri_glmfit-sim. The issue as you explained was that the original annotation was not being cleared. You suggested a short-term as well as a long-term fix.
The temporary fix was deleting "-annot aparc" from mri_glmfit-sim. As a more permanent fix, you emailed me a modified copy of mri_surfcluster.
- My question is when re-running mri_glmfit-sim with the new mri_surfcluster do I still need to delete "-annot aparc" from mri_glmfit-sim? In other words, are both of these needed or will using the modified mri_surfcluster alone solve the problem?
using the new mri_surfcluster will solve the problem
- Another question: Is there any way to convert the previous "bad" .annot file to "good" .annot file so we can view it without having to run mri_glmfit-sim again? We ran it with 5000 iterations and it took over 10 days. If there is a way we can get the same results without having to re-run it, I would very much like to know about it.
Just run it with "--no-sim" and it will just redo the clustering.
dogu
Thank you for your help so far. Devdutta
On Tue, Mar 17, 2009 at 12:26 PM, Devdutta W <devdutta.w@gmail.com mailto:devdutta.w@gmail.com> wrote:
Hi Doug, Thanks for your help. A followup question to your last email: the short term fix that you mentioned earlier, do I need to make that fix in addition to copying the new mri_surfcluster file in the binary dir? Or has that been fixed already? Thanks, Devdutta On Mon, Mar 9, 2009 at 4:41 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote: Try copying this: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/mri_surfcluster into your $FREESURFER_HOME/bin directory (make a backup first), and then re-run. doug Devdutta W wrote: We are using Red Hat Enterprise Linux 5. Thanks for your help. I'll try that out. On Mon, Mar 9, 2009 at 3:02 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>> wrote: I think I've found the problem in that the original annotation at a vertex is not being cleared if a cluster is not there. The short term fix is to delete " --annot aparc" from mri_glmfit-sim at line 353. I'll email when I have a longer term fix. What platform are you using? doug Devdutta W wrote: When I load mc-z.abs.2.sig.cluster.mgh, if the "Draw filled labels" button is toggled on, it looks no different from before. Toggling the "show overlay" button doesn't do anything. If the "Draw Outlined labels" button is toggled on, the outlines of the colored regions can be seen (including the outlines of the clusters). This time, toggling the overlay button on-off shows the clusters. On Mon, Mar 9, 2009 at 2:12 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>>> wrote: /what does it look like when you load mc-z.abs.2.sig.cluster.mgh as an overlay?/ Devdutta W wrote: Nick, Sita, The average subject is created from our data, perhaps that's why it looks different. We were having problems getting freesurfer to recognize our average subject. So we linked fsaverage to our average subject. I didn't see a button (or a tool) to toggle on-off the annotation. The overlay button is there but toggling it doesn't do anything. There indeed are clusters mixed in with the annotations. But from the documenation, it seems that the command we are using is /supposed/ to show clusters (but only the clusters, not the the annotations). And that is the question I am trying to get answered. Here is the command we are using: /tksurfer fsaverage lh inflated -annot mc-z.abs.2.sig.ocn.annot -fthresh 2 -curv -gray/ I am running this command from the appropriate folder and fsaverage is linked to our average subject. We are using the following documentation: https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis?highlight=(mri <https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis?highlight=%28mri> <https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis?highlight=%28mri> <https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis?highlight=%28mri> <https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis?highlight=%28mri>\_glmfit\-sim) Thanks, Devdutta On Mon, Mar 9, 2009 at 12:09 PM, Nick Schmansky <nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu>> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu>>> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu>> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu>>>>> wrote: Devdutta, What version of freesurfer is being used? The fsaverage subject you show doesnt look right. Also, I see what appear to be clusters mixed in with the annotations. The summary files output by the multiple correction step will tell you how many and the size of the clusters. You should be able to toggle on-off both the annotation and the overlay independently. I am cc'ing Sita, as she may be able to help you on this (I will have limited email contact this week). Nick On Mon, 2009-03-09 at 11:21 -0500, Devdutta W wrote: > Forgot to attach the images, again. > > > On Mon, Mar 9, 2009 at 11:20 AM, Devdutta W <devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>>>>> > wrote: > Hi Nick, > Can you shed some light on why this might be happening? I > will attach the two picture again in case you weren't able to > view them last time. I tried what I thought you meant by > "turn off annotation display" and I have stated what happened > in my previous email (please see below) > > * If I don't include the -annot option in tksurfer it > only displays the fsaverage, naturally. > * If I toggle off the button that says "show labels", it > looks like it is just the fsaverage as well. > * If I toggle on the "Draw outlined labels" button it > shows the same regions as before except they are > merely outlined, not filled. > Any help you can give me will be appreciated. > > Thanks, > Devdutta > > > On Wed, Mar 4, 2009 at 10:36 AM, Devdutta W > <devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>>>>> wrote: > Nick, > I confess, I am not exactly sure what you mean by > "turn off annotation display". I tried a few > things. > > * If I don't include the -annot option in > tksurfer it only displays the fsaverage, > naturally. > * If I toggle off the button that says "show > labels", it looks like it is just the > fsaverage as well. > * If I toggle on the "Draw outlined labels" > button it shows the same regions as before > except they are merely outlined, not filled. > These were the only ones that I could think of. Is > that what you meant or was it something else that I > missed? > > Thanks, > Devdutta > > > On Tue, Mar 3, 2009 at 7:12 PM, Nick Schmansky > <nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu>> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu>>> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu>> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu> <mailto:nicks@nmr.mgh.harvard.edu <mailto:nicks@nmr.mgh.harvard.edu>>>>> wrote: > What is displayed if you turn off the > annotation display? > > The threshold in the simulation is used to > determine whether clusters > are found or not. A threshold of 2 indicates > a p-value of 0.01 > (1/(10^2), 3 is 0.001, etc. The threshold > referred to in the tksurfer > display is just for color thresholding. > > Nick > > > On Tue, 2009-03-03 at 17:17 -0600, Devdutta W > wrote: > > Forgot the attachments earlier. Here they > are now > > > > On Tue, Mar 3, 2009 at 5:11 PM, Devdutta W > <devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com> <mailto:devdutta.w@gmail.com <mailto:devdutta.w@gmail.com>>>>> > > wrote: > > Hi Freesurfers, > > More questions about this analysis. > > 1) Referring to the documentation, > we used the following > > command: > > tksurfer fsaverage lh inflated - > annot mc- > > z.abs.2.sig.ocn.annot -fthresh 2 - > curv -gray > > > > The documentation says we should see > an image as shown in > > Figure 1 (attached). But what we > see is Figure 2 (also > > attached). Can somebody explain > this? > > > > 2) When running the simulation does > it matter what threshold > > we use? For example, can we use a > threshold of 2 or 3 > > (instead of 4 as shown in > documentation) when running > > mri_glmfit-sim? And do we have to > use the same threshold > > while displaying the clusters after? > > > > Thanks for any help. > > > > Devdutta > > > > > > _______________________________________________ > > Freesurfer mailing list > > Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>>> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>>>> > > > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > > > > ------------------------------------------------------------------------ _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer