Hi guys,
I (and probably others) have noticed the aparc.a2009s label naming convention has changed in recons starting from 6.0. “_and_” has become “&”, such as in lh.G&S_subcentral. It is fixable, but it may become a bit confusing for those who use fsaverage vs. individually segmented labels in a study that uses the different versions of 6.0 and beyond (e.g., those interested in using the subpmillmeter recon option). And & is also a special character, too.
It is completely fixable, of course, but I just thought to bring this up, because it might cause confusion in some users. (For me, it was just a “why the xxxx does not my mris_divide_parcellation splittable” moment.)
Thankfully,
Jyrki
External Email - Use Caution
Yes, this has indeed also caused me some confusion.
I took the label names from fsaverage (and thus from my v6 installation), parsed ABIDE structural data (5.x) and wondered why there was no data in there for most atlas labels. I was quite sure it was an issue with the fact that '%' is a special character at first (as a programmer I smell trouble immediately when seeing special characters), but at some point understood that the labels had simple been renamed. oO
On August 14, 2019 at 9:47 PM Jyrki Ahveninen jyrki@nmr.mgh.harvard.edu wrote:
Hi guys,
I (and probably others) have noticed the aparc.a2009s label naming convention has changed in recons starting from 6.0. “_and_” has become “&”, such as in lh.G&S_subcentral. It is fixable, but it may become a bit confusing for those who use fsaverage vs. individually segmented labels in a study that uses the different versions of 6.0 and beyond (e.g., those interested in using the subpmillmeter recon option). And & is also a special character, too.
It is completely fixable, of course, but I just thought to bring this up, because it might cause confusion in some users. (For me, it was just a “why the xxxx does not my mris_divide_parcellation splittable” moment.)
Thankfully,
Jyrki _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
External Email - Use Caution
Another interesting thing I noticed, which may be of interest to anybody parsing label data: in the ABIDE data (5.x), labels without any assigned vertices are not listed at all for subjects.
So if you want to find all label names of an atlas, you cannot just take a random subject and read them from its ?h.aparc.a2009s.annot file: the subject you chose may be lacking some of them. (That's why I went for fsaverage, which was a bad idea as well.)
Tim
On August 15, 2019 at 9:37 AM Tim Schäfer ts+ml@rcmd.org wrote:
External Email - Use CautionYes, this has indeed also caused me some confusion.
I took the label names from fsaverage (and thus from my v6 installation), parsed ABIDE structural data (5.x) and wondered why there was no data in there for most atlas labels. I was quite sure it was an issue with the fact that '%' is a special character at first (as a programmer I smell trouble immediately when seeing special characters), but at some point understood that the labels had simple been renamed. oO
On August 14, 2019 at 9:47 PM Jyrki Ahveninen jyrki@nmr.mgh.harvard.edu wrote:
Hi guys,
I (and probably others) have noticed the aparc.a2009s label naming convention has changed in recons starting from 6.0. “_and_” has become “&”, such as in lh.G&S_subcentral. It is fixable, but it may become a bit confusing for those who use fsaverage vs. individually segmented labels in a study that uses the different versions of 6.0 and beyond (e.g., those interested in using the subpmillmeter recon option). And & is also a special character, too.
It is completely fixable, of course, but I just thought to bring this up, because it might cause confusion in some users. (For me, it was just a “why the xxxx does not my mris_divide_parcellation splittable” moment.)
Thankfully,
Jyrki _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Hi Tim
I think that is the nature of the parcellation - there are some small folds that don't occur in all subjects.
cheers Bruce On Thu, 15 Aug 2019, Tim Schäfer wrote:
External Email - Use CautionAnother interesting thing I noticed, which may be of interest to anybody parsing label data: in the ABIDE data (5.x), labels without any assigned vertices are not listed at all for subjects.
So if you want to find all label names of an atlas, you cannot just take a random subject and read them from its ?h.aparc.a2009s.annot file: the subject you chose may be lacking some of them. (That's why I went for fsaverage, which was a bad idea as well.)
Tim
On August 15, 2019 at 9:37 AM Tim Schäfer ts+ml@rcmd.org wrote:
External Email - Use CautionYes, this has indeed also caused me some confusion.
I took the label names from fsaverage (and thus from my v6 installation), parsed ABIDE structural data (5.x) and wondered why there was no data in there for most atlas labels. I was quite sure it was an issue with the fact that '%' is a special character at first (as a programmer I smell trouble immediately when seeing special characters), but at some point understood that the labels had simple been renamed. oO
On August 14, 2019 at 9:47 PM Jyrki Ahveninen jyrki@nmr.mgh.harvard.edu wrote:
Hi guys,
I (and probably others) have noticed the aparc.a2009s label naming convention has changed in recons starting from 6.0. “_and_” has become “&”, such as in lh.G&S_subcentral. It is fixable, but it may become a bit confusing for those who use fsaverage vs. individually segmented labels in a study that uses the different versions of 6.0 and beyond (e.g., those interested in using the subpmillmeter recon option). And & is also a special character, too.
It is completely fixable, of course, but I just thought to bring this up, because it might cause confusion in some users. (For me, it was just a “why the xxxx does not my mris_divide_parcellation splittable” moment.)
Thankfully,
Jyrki _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
External Email - Use Caution
Hi Bruce,
yes, but to clarify, I was referring to something else:
If a subject is missing a fold of an atlas, I somehow expected there would still be an entry for the fold in the atlas stats file (which would list 0 vertices and NaNs for the mean thickness and other measurements), but the whole entry is missing instead.
That's fine, you just have to be aware of it.
Tim
On August 16, 2019 at 4:02 PM Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Tim
I think that is the nature of the parcellation - there are some small folds that don't occur in all subjects.
cheers Bruce On Thu, 15 Aug 2019, Tim Schäfer wrote:
External Email - Use CautionAnother interesting thing I noticed, which may be of interest to anybody parsing label data: in the ABIDE data (5.x), labels without any assigned vertices are not listed at all for subjects.
So if you want to find all label names of an atlas, you cannot just take a random subject and read them from its ?h.aparc.a2009s.annot file: the subject you chose may be lacking some of them. (That's why I went for fsaverage, which was a bad idea as well.)
Tim
On August 15, 2019 at 9:37 AM Tim Schäfer ts+ml@rcmd.org wrote:
External Email - Use CautionYes, this has indeed also caused me some confusion.
I took the label names from fsaverage (and thus from my v6 installation), parsed ABIDE structural data (5.x) and wondered why there was no data in there for most atlas labels. I was quite sure it was an issue with the fact that '%' is a special character at first (as a programmer I smell trouble immediately when seeing special characters), but at some point understood that the labels had simple been renamed. oO
On August 14, 2019 at 9:47 PM Jyrki Ahveninen jyrki@nmr.mgh.harvard.edu wrote:
Hi guys,
I (and probably others) have noticed the aparc.a2009s label naming convention has changed in recons starting from 6.0. “_and_” has become “&”, such as in lh.G&S_subcentral. It is fixable, but it may become a bit confusing for those who use fsaverage vs. individually segmented labels in a study that uses the different versions of 6.0 and beyond (e.g., those interested in using the subpmillmeter recon option). And & is also a special character, too.
It is completely fixable, of course, but I just thought to bring this up, because it might cause confusion in some users. (For me, it was just a “why the xxxx does not my mris_divide_parcellation splittable” moment.)
Thankfully,
Jyrki _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
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
I see, and I agree that would be better cheers Bruce
On Sat, 17 Aug 2019, Tim Schäfer wrote:
External Email - Use CautionHi Bruce,
yes, but to clarify, I was referring to something else:
If a subject is missing a fold of an atlas, I somehow expected there would still be an entry for the fold in the atlas stats file (which would list 0 vertices and NaNs for the mean thickness and other measurements), but the whole entry is missing instead.
That's fine, you just have to be aware of it.
Tim
On August 16, 2019 at 4:02 PM Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Tim
I think that is the nature of the parcellation - there are some small folds that don't occur in all subjects.
cheers Bruce On Thu, 15 Aug 2019, Tim Schäfer wrote:
External Email - Use CautionAnother interesting thing I noticed, which may be of interest to anybody parsing label data: in the ABIDE data (5.x), labels without any assigned vertices are not listed at all for subjects.
So if you want to find all label names of an atlas, you cannot just take a random subject and read them from its ?h.aparc.a2009s.annot file: the subject you chose may be lacking some of them. (That's why I went for fsaverage, which was a bad idea as well.)
Tim
On August 15, 2019 at 9:37 AM Tim Schäfer ts+ml@rcmd.org wrote:
External Email - Use CautionYes, this has indeed also caused me some confusion.
I took the label names from fsaverage (and thus from my v6 installation), parsed ABIDE structural data (5.x) and wondered why there was no data in there for most atlas labels. I was quite sure it was an issue with the fact that '%' is a special character at first (as a programmer I smell trouble immediately when seeing special characters), but at some point understood that the labels had simple been renamed. oO
On August 14, 2019 at 9:47 PM Jyrki Ahveninen jyrki@nmr.mgh.harvard.edu wrote:
Hi guys,
I (and probably others) have noticed the aparc.a2009s label naming convention has changed in recons starting from 6.0. “_and_” has become “&”, such as in lh.G&S_subcentral. It is fixable, but it may become a bit confusing for those who use fsaverage vs. individually segmented labels in a study that uses the different versions of 6.0 and beyond (e.g., those interested in using the subpmillmeter recon option). And & is also a special character, too.
It is completely fixable, of course, but I just thought to bring this up, because it might cause confusion in some users. (For me, it was just a “why the xxxx does not my mris_divide_parcellation splittable” moment.)
Thankfully,
Jyrki _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
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
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Jyrki Ahveninen -
Tim Schäfer