I do now have F-tests working in MRtrix3 (I think ). However it lives symbiotically with a suite of other changes overhauling the statistics code; these reside in a private branch that is currently undergoing testing, both internally and with a couple of collaborators, and the features can’t really be separated from one another due to the shear complexity and inter-dependency of the changes.
I’m still trying to figure out the right balance between verifying that the code is doing as it should, giving collaborators the opportunity to perform studies utilising these capabilities with my oversight, actually getting some of my own work done (), but acknowledging that many members of the wider community are eager to get their hands on some of these capabilities as soon as possible. So while I don’t have it ready for the upcoming
3.0_RC3 release candidate update, I’ll hopefully get it all together for the subsequent release.