 |
 |
Hi Susanna, Some differences would be expected given that the data is being permuted. Hopefully real effects in the data wouldn’t disappear due to permutation effects. Would you be able to share your design files (*.con, *.mat, *.grp, *.fsf)? You can upload to Oxford’s file sharing service with this link: https://oxfile.ox.ac.uk/oxfile/work/extBox?id=738200123033B05CFC Cheers, Taylor Hanayik Taylor Hanayik PhD Analysis Research Software Engineer FMRIB, John Radcliffe Hospital University of Oxford [log in to unmask] > On 3 Jun 2019, at 15:04, Susanna Gobbi <[log in to unmask]> wrote: > > Dear FSL users, > > I was using FSL both on the cloud and locally on my machine, it is a Ubuntu 16.04.5 LTS xenial in both cases. I had also the same version of FSL 6.0, but due to an issue with an FSL function in the cloud (eddy_openmp), I had to remove and to re-install FSL, I used fslinstaller.py. > After this re-installation, I run again some old second level analysis with FSL randomise and I have noticed that the function now seems to perform faster for 500 permutations than before. > When I checked my second level results they looked significantly different from the one obtained before re-installing FSL. > Do you know if recently they changed something in the algorithm of the randomise (version v2.9)? Have you ever experienced something similar, maybe using other FSL functions before and after re-installing FSL? > > I am really thankful for your help and suggestions, > > Best wishes, > Susanna > > ######################################################################## > > To unsubscribe from the FSL list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ######################################################################## To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1