Correlation files

Hi all

First of all I have to say that I'm really liking the DPARSF extension, it's very cool.

I've used DPARSF to calculate FC using 4 user-defined ROIs. In the textfile "ROISignals_subjectname.txt" I find the extracted time courses for my 4 ROIs in 4 columns, each row representing my number of time points. However, when looking at the "ROICCorrelation_subjectname.txt" I only find 4 columns and 4 rows. I'm not really sure how to read this data and use it in further analysis.

If someone could give me some pointers I would be deeply grateful!

/Eva

Forums:

Thanks, Eva!

"ROICCorrelation_subjectname.txt" I only find 4 columns and 4 rows.

Those are Pearson's correlation coefficients. Row 1 Column 2 means the correlation between the 1st ROI and the 2nd ROI. Row 2 Column 4 means the correlation between the 2nd ROI and the 4th ROI.

Best,

Chao-Gan

Thank you for your quick reply Chao-Gan! Now it all makes sense, I was stuck on the idea that rows were timepoints.

Thanks!

/Eva

 

Hello again

I've ran into some more problems that I would really appreciate som help with.

I now have a textfile for each subject that contains the (Z transformed) correlation coefficients for 4 different ROIs.  I want to check the correlations for my ROIs on a group level but I have no idea how to do that.

Thankful for help

/Eva

Hi Eva,

It depends on how much you want to put into matlab commands.

1. For matlab, you may do: 1) load 4x4 matrices and concatenate them into 4x4xN matrix; 2) use matlab to perform group comparison, e.g., ttest or ttest2.

2. If you don't want to use matlab, you can copy the corresponding number into excel, then use excel or SPSS to do group analysis.

Best,

Chao-Gan