Connectometry analysis: Study Region, possible bug?

8 views
Skip to first unread message

Aleksi Sihvonen

unread,
Apr 23, 2021, 7:03:43 AM4/23/21
to DSI Studio
Hi Frank,

I have ran group connectometry analysis and found significant results in tracts A and B. I tried to replicate these results using tracts A and B as ROIs in Study Region -option following the manual. Just to learn this method too knowing that there should be significant results. However:

1. Using tracts A and B as ROI, I cannot find any results, despite increased permutations to 8000+. However, using tracts A and B as SEED, this works. According to the manual, the ROI approach should work too?
2. In the end of the analysis, the program crashes and fails to output the report. It outputs pos, neg and t-statistics files as well as a text file fdr_dist_values in the folder, but should it output the normal report too? I have the latest build.

Thank you for your time and support again.

Best,
Aleksi

Frank Yeh

unread,
Apr 23, 2021, 10:30:45 AM4/23/21
to dsi-s...@googlegroups.com
The FDR is calculated by the ratio of findings between permuted and
non-permuted demographics. Adding more than one region will heavily
restrict how the algorithm utilizes nonpremuted demographics for FDR
calculations.
Usually, I used only one ROI with even more permutations like 10,000
so that the FDR calculation can converge.

Hope this helps.
Frank
> --
> You received this message because you are subscribed to the Google Groups "DSI Studio" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to dsi-studio+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/c2f77765-26c2-4684-8df4-41150dfb6ffan%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages