A desire for more tedana options

AFNI Message Board
1

Hey AFNI folks,

The tedana/multiecho integration in afni proc is extraordinary, but I have a simple (haha) request. Currently, afni_proc provides a nice litany of options for combining the data, however tedana outputs more combined/denoised files than afni_proc has built-in access to. For example, if I wanted to use the dn_ts_T1c or the tsoc_T1c file produced by tedana, I would have to modify the afni_proc output script for each subject. It would be great to add in perhaps one additional option, something like -combine_dset_name or -combine_tedana_dset, that would allow the user to input the filename of the tedana output file themselves.

I prefer this idea over more -combine_method options, as this would somewhat future proof afni_proc if new combination methods are integrated into various tedana instances.

I am also happy to try and add this myself, and do a pull request, but I’m not sure where the modification would be made on the github.

Thanks again for all the awesome!

2

That sounds reasonable, but I have no idea what those
files are. Are they coming from the updated software
from Elizabeth?

  • rick
3

I believe they are present in that version but they are also produced by a more recent version that the one included with AFNI (newer version is 3.2, I believe). That was the latest version on bitbucket, and was migrated over to github. There is also the high-kappa (hik_ts) timeseries that could be desired and used, though may need some modification (mean zero).

Thanks for looking into it!