[color=#000000]Hi David,[/color]
[color=#000000]The denoised timeseries are only saved into a CONN's internal .matc format (which makes reading data from individual slices faster to make the GUI more responsive). If you want CONN to create a NIFTI file with the denosied timeseries, simply check the option that reads "[i]create confound-corrected timeseries[/i]" in the [i]Setup.Options [/i]tab before running the Denoising step (that will create a file in the same directory as your original functional data with the filename prepended by 'd' and containing the denoised timeseries)[/color]
And regarding units, yes, you are exactly right, the seed-to-voxel maps are typically (when using all default settings) Fisher-transformed correlation coefficients (these are often also called z-coefficients although they are not really scaled to unit-variance), while the ICA spatial components are in standard unit-variance z-score units
[color=#000000]Hope this helps[/color]
[color=#000000]Alfonso[/color]
[i]Originally posted by David Fischer:[/i][quote]Hello,
I'm trying to find the output file which contains the time series for each voxel, after denoising. In the 1st level analysis in the GUI, I have the option of clicking on a given voxel for a given subject, and CONN will display the time series for that voxel, for that subject. I'm looking for that file which that GUI function is based on (or looking for another way of extracting the time series on a voxel-by-voxel basis).
From a previous thread, it would seem that file would be: niftiDATA_Subjectxxx_Conditionxxx.nii
However, I only see niftiNORMS_Subjectxxx_Conditionxxx.nii in the results/preprocessing directory.
And an unrelated question: I gather that the units for displaying a single subject seed-to-voxel beta map are Z scores. What are the units for displaying a single subject ICA component? Is it still a Z score or something different?
Thank you!!
David[/quote]
[color=#000000]The denoised timeseries are only saved into a CONN's internal .matc format (which makes reading data from individual slices faster to make the GUI more responsive). If you want CONN to create a NIFTI file with the denosied timeseries, simply check the option that reads "[i]create confound-corrected timeseries[/i]" in the [i]Setup.Options [/i]tab before running the Denoising step (that will create a file in the same directory as your original functional data with the filename prepended by 'd' and containing the denoised timeseries)[/color]
And regarding units, yes, you are exactly right, the seed-to-voxel maps are typically (when using all default settings) Fisher-transformed correlation coefficients (these are often also called z-coefficients although they are not really scaled to unit-variance), while the ICA spatial components are in standard unit-variance z-score units
[color=#000000]Hope this helps[/color]
[color=#000000]Alfonso[/color]
[i]Originally posted by David Fischer:[/i][quote]Hello,
I'm trying to find the output file which contains the time series for each voxel, after denoising. In the 1st level analysis in the GUI, I have the option of clicking on a given voxel for a given subject, and CONN will display the time series for that voxel, for that subject. I'm looking for that file which that GUI function is based on (or looking for another way of extracting the time series on a voxel-by-voxel basis).
From a previous thread, it would seem that file would be: niftiDATA_Subjectxxx_Conditionxxx.nii
However, I only see niftiNORMS_Subjectxxx_Conditionxxx.nii in the results/preprocessing directory.
And an unrelated question: I gather that the units for displaying a single subject seed-to-voxel beta map are Z scores. What are the units for displaying a single subject ICA component? Is it still a Z score or something different?
Thank you!!
David[/quote]