Mean FA line of the group differs between the group comparison and the correlation analysis

[warddef]

Dear Jakob,

I have a question regarding plots which come out of plot tractometry results, as I understood from the tractometry documentation, the two lines in the group comparison represent the mean FA value of the two groups. As I also understood from issue #145, the blue graph in the correlation analysis is still the mean FA of the group. However when comparing the graphs from the group analysis

to the graphs from the correlation analysis

you can see that there is little to no over
lap between the FA values in these different plots. While I realise that this can be due to scale differences between the group comparison and the correlation graphs, the scale does not differ enough to warrant such large differences. Since I thought that this was a simple scale problem (because the pval, significant regions and alpha FWE stayed the same) I tried troubleshooting it myself by either restricting the scales using --range '0.1 0.6' or by adding ax.set_ylim(0, 0.6) to the source code of plot tractometry results. Neither of these solved the issue. Comparing one bundle at a time instead of twelve at once did also not make a difference. Do you have any suggestions on how I can solve this, any input would be greatly appreciated. in attachment the input txt files, and code I use for the analysis.

analysis final .txt
subjects.txt

subjects_correlation_SRS_total_NT.txt
subjects_correlation_SRS_total_PT.txt

Kind regards,

Ward

[wasserth]

That's a good question. I do not have a direct answer. I will have to check myself on a dataset.

[warddef]

have you already had a chance to look at it ? if it helps I can share the data I used.