I didn't previously realize this was on the hcp-cifti list, which is in fact intended for technical (programming, etc) questions about the cifti format. I am moving this to the hcp-users list so that others can see and respond to it.
If your values result from using the Harvard-Oxford atlas areas to make averages, then a parcellated cifti file could be appropriate, and would preserve the exact data values, rather than turning them into node sizes. Unfortunately, it appears to be a volume-based atlas, which has some drawbacks. If these correlations are cortical, we would recommend that you use surface-based analysis instead, and use a surface atlas for cortex (such as the HCP MMP1.0). If you want to display them as spheres anyway, the only obvious way I know of to do this is to enter "foci" mode in wb_view, and enter them manually. I believe you will need to have surfaces loaded for this to work. Note that foci may change the locations (coordinates) they display in if you use different surfaces to display them later. wb_command is not intended to be run by double-clicking. It is a command-line utility, that is, you first open a terminal window, and then run it by typing in "wb_command" (if your $PATH is set correctly). It does not open a window, its only output besides writing files is putting text into a terminal window. You can explore its capabilities here, which are basically webpages made by capturing this text output for wb_command's help and information options: https://www.humanconnectome.org/software/workbench-command Tim On Thu, Mar 29, 2018 at 9:57 AM, Jeena Thomas <jeena.th...@gmail.com> wrote: > Hi Tim, > > Thank you for this information. I currently have a Harvard Oxford atlas > nii.gz file and I have various points of correlations, indicated by the > .node files, determined statistically. I want to be able to map those > points onto the Harvard Oxford Atlas. I was trying to use workbench to do > so. > > Also, considering that Workbench is on my Macbook, I cannot seem to open > the wb_command icon. When I click to open it, even if I right click and > select 'open', it will not open. Do you have any insight into this? > > > Let me know what you think! > > Thanks, > > Jeena > > On Mon, Mar 26, 2018 at 5:05 PM, Timothy Coalson <tsc...@mst.edu> wrote: > >> Surface atlases have a different nature to volume atlases, in that >> surface atlases are actually maps of features in spherical coordinates - >> anatomical coordinates are optional and often not used in registration (a >> sulcal-depth-like measure is used instead when anatomical registration is >> desired). I am not sure what you are trying to do, but converting a >> volumetric atlas template to a .surf.gii (geometry) file is a very uncommon >> task, and doesn't have a particularly clean solution that I am aware of. >> >> The general surface analysis pipeline is to first segment every subject's >> structural images, make surfaces for each subject, map each subject's data >> onto their own surfaces, register the subjects' surfaces to an atlas, >> resample each subject's surfaces and surface data to use the atlas >> topology, and then do cross-subject statistics. Based on this, can you >> tell us what you are trying to do and where it diverges from this? >> >> Tim >> >> >> On Mon, Mar 26, 2018 at 2:51 PM, Jeena Thomas <jeena.th...@gmail.com> >> wrote: >> >>> Also, speaking of which, if you have any recommendations for converting >>> an atlas that I have in nii. and nii.gz into a surf.gii, that would be >>> really great! >>> >>> >>> Thanks, >>> >>> Jeena >>> >>> On Mon, Mar 26, 2018 at 3:35 PM, Jeena Thomas <jeena.th...@gmail.com> >>> wrote: >>> >>>> The columns are as follows: >>>> [x] [y] [z] [color] [node size: strength of correlation] [brain anatomy >>>> label] >>>> >>>> Note, though that the [brain anatomy label] is a number that currently >>>> does not correlate to actual anatomical regions because I do not have >>>> specified regions for the atlas I am using (Harvard Oxford Atlas). The >>>> fourth column that indicates 'color' indicates positive or negative >>>> correlation. >>>> >>>> >>>> Thank you for your time and patience, >>>> >>>> Jeena >>>> >>>> >>>> >>>> >>>> On Mon, Mar 26, 2018 at 2:10 PM, Glasser, Matthew <glass...@wustl.edu> >>>> wrote: >>>> >>>>> In that case you need to tell me what each column in the file >>>>> represents and I can suggest what NIFTI/GIFTI/CIFTI formats you should >>>>> use. >>>>> >>>>> Peace, >>>>> >>>>> Matt. >>>>> >>>>> From: Jeena Thomas <jeena.th...@gmail.com> >>>>> Date: Monday, March 26, 2018 at 1:07 PM >>>>> >>>>> To: Matt Glasser <glass...@wustl.edu> >>>>> Cc: "hcp-ci...@humanconnectome.org" <hcp-ci...@humanconnectome.org> >>>>> Subject: Re: [HCP-CIFTI] nodes files to CIFTI files >>>>> >>>>> No software makes it. I just made it with a script. The node file >>>>> essentially has 6 columns, three of which relate to coordinates that can >>>>> be >>>>> mapped onto a brain. >>>>> >>>>> On Mon, Mar 26, 2018 at 1:57 PM, Glasser, Matthew <glass...@wustl.edu> >>>>> wrote: >>>>> >>>>>> That Isn’t a standard neuroimaging file type that I have heard of. >>>>>> What matlab software makes them? >>>>>> >>>>>> Peace, >>>>>> >>>>>> Matt. >>>>>> >>>>>> From: Jeena Thomas <jeena.th...@gmail.com> >>>>>> Date: Monday, March 26, 2018 at 12:55 PM >>>>>> To: Matt Glasser <glass...@wustl.edu> >>>>>> Cc: "hcp-ci...@humanconnectome.org" <hcp-ci...@humanconnectome.org> >>>>>> Subject: Re: [HCP-CIFTI] nodes files to CIFTI files >>>>>> >>>>>> Hi Matt, >>>>>> >>>>>> I have been able to create .node files that represent points of >>>>>> correlation from connectivity analysis. So I was wondering how to map >>>>>> those >>>>>> points onto visual programs like Workbench. I thought perhaps converting >>>>>> a >>>>>> .node file to a CIfTI file, i.e. a dlabel.nii file, would do the trick. >>>>>> >>>>>> >>>>>> Let me know what you think, >>>>>> >>>>>> Jeena >>>>>> >>>>>> On Mar 26, 2018, at 1:09 PM, Glasser, Matthew <glass...@wustl.edu> >>>>>> wrote: >>>>>> >>>>>> I don’t know what a node file is. Can you give more details? >>>>>> >>>>>> Peace, >>>>>> >>>>>> Matt. >>>>>> >>>>>> From: <hcp-cifti-boun...@humanconnectome.org> on behalf of Jeena >>>>>> Thomas <jeena.th...@gmail.com> >>>>>> Date: Monday, March 26, 2018 at 11:25 AM >>>>>> To: "hcp-ci...@humanconnectome.org" <hcp-ci...@humanconnectome.org> >>>>>> Subject: [HCP-CIFTI] nodes files to CIFTI files >>>>>> >>>>>> To Whom It May Concern: >>>>>> >>>>>> I was wondering if there was a way to convert node files (made on >>>>>> Matlab) to CIFTI files, especially dlabel.nii files. >>>>>> >>>>>> >>>>>> Thank you, >>>>>> >>>>>> -- >>>>>> Jeena Thomas >>>>>> _______________________________________________ >>>>>> HCP-CIFTI mailing list >>>>>> hcp-ci...@humanconnectome.org >>>>>> http://lists.humanconnectome.org/mailman/listinfo/hcp-cifti >>>>>> >>>>>> >>>>>> >>>>> >>>>> >>>>> -- >>>>> Jeena Thomas >>>>> >>>> >>>> >>>> >>>> -- >>>> Jeena Thomas >>>> >>> >>> >>> >>> -- >>> Jeena Thomas >>> >>> _______________________________________________ >>> HCP-CIFTI mailing list >>> hcp-ci...@humanconnectome.org >>> http://lists.humanconnectome.org/mailman/listinfo/hcp-cifti >>> >> >> > > > -- > Jeena Thomas > _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users
