Taylor Salo

>Since there is little use of individual connectivity estimates, it would be interesting to see how it impacts group analyses, e.g. in a classification task, as we do it here...

That's my understanding as well. However, it seems (and should of course be empirically validated) that the number of low-motion volumes will impact the variance, rather than the correlation estimate,...

>I had begun to refactor and add docstrings and unit tests as a branch of the [xDF repository](https://github.com/griegner/xDF/tree/gr), so transitioning it to Nilearn or PyPi would be a doable next...

Part of this can involve breaking up the example walkthrough, which is quite long. The CuBIDS commands we have are: - validate - bids-version - sidecar-merge - group - apply...

So is the idea something like `cubids mv /path/to/current_file /path/to/new_file` and it would go through the associated files and fix them as well? I think that's worth tackling before the...

It would be cool if this supported directories as well. E.g., `cubids mv sub-01 sub-02` would rename all of the files as well.

This is going to introduce a lot of merge conflicts with any active PRs so I plan to tackle this at the very end of the code sprint.

Here's what our coverage timeline looks like on CodeCov: 

I haven't removed `read_nifti_sidecar`, which is a big barrier to using the inheritance principle in DWI datasets.

No, but I'm seeing the same failure in #766. I think it might be an issue with DiPy.