Interpretations of CWAS results

Submitted by louislin28 on

Hi,

Thank you for your kind and tremendous help, we finally had the processes of CWAS calculation complete. I tried to use REST viewer and DPABI viewer to present the results of p-value- or pseudo-F- maps, but failed. 

What I want to do is like  the method described in Shehzad et al. 2014: "The pseudo-F statistic from the original data was then referred to this simulated distribution to obtain a p-value. We controlled for family-wise error rate using Gaussian Random Field (GRF) theory (voxel threshold of Z > 1.65, cluster size threshold p < 0.05; note the one-tailed threshold was employed due to the unidirectional nature of the F-test)".

I put the pseudo-F-map (and p-value map following failure to present pseudo-F-map) into the overlay in REST viewer. Then I chose GRF correction, and fed in my custom-made mask, put the threshold according to the Shehzad work. Then the error message showed:

Error in rest_sliceviewer>Overlay_Misc (line 3928)
                    [Data_Corrected, ClusterSize, Header]=rest_GRF_Threshold(AConfig.Overlay.Volume,...

Error in rest_sliceviewer (line 678)
            REST_SliceViewer_Cfg.Config(theCardinal) =Overlay_Misc(REST_SliceViewer_Cfg.Config(theCardinal));
 
Error while evaluating uicontrol Callback

 

How can I use the pseudo-F and p-value maps yielded from CWAS calculations in DPARSF to present the results?

Thank you very very much

 

YAN Chao-Gan

Fri, 10/23/2015 - 06:48

Hi,

You already have the p-value map. In Shehzad paper, they converted p to z for GRF correction if I remember correctly. GRF correction on the pseudo-F map is not appropriate.

You can try http://rfmri.org/Programs_YAN #9 for converting p maps to z maps.

Best,

Chao-Gan