Dear Experts,

We putted "A_z_p.nii.gz" and "B_z_p.nii.gz" in the multimodal analyses for exploring the relationship of A and B, and the probability was set as 0.0025.
The brain regions reported in the multimodal results (A_neg_B_neg_p_0.00250_0.00025_10.nii.gz) have appeared in the results of comparing A and controls (A_z_p_0.00500_1.000_10.nii.gz), but not in the results of comparing B and controls (B_z_p_0.00500_1.000_10.nii.gz).
Is this normal? Do we need to use a stricter probability?
Thank you very much in advance!

All the best,
You.

Dear You, 

Do you refer that the results of the multimodal had been saved in the folder of the A meta-analysis?? It is that what you mean, don't worry about it. That means that you set that folder to save the ouput but it has nothing to do with the analysis. When performing the multimodal analysis, you can select where you want to save the output, by default is the folder open in SDM.

Kind regards, 
Lydia

Dear Lydia,

Thank you for your response!

I'm still wondering that if it is reasonable that a brain region was reported in the combined multimodal results but not reported in previous separate analysis. For example, we explored the disorder A and disorder B. Patients with disorder A showed abnormality in prefrontal cortex, but disorder B does not report changes in prefrontal cortex or adjacent brain regions. Then, we performed multimodal analysis between disorder A and disorder B. The results of multimodal analysis exhibited abnormality in prefrontal cortex. This result makes us confused. Why results not reported in disease B appeared in the multimodal results. I’m not sure whether we performed the analysis correctly or do you have any other suggestions please let us know.

Your help and advice would be much appreciated!

You.

Dear You, 

One question, the results in prefrontal cortex doesn't appear in the corrected or uncorrected results?. One possible explanation is that the region is close to significance (imaginge a threshold of 0.05, and it has 0.051). So, when you pool all, it takes this into account considering that is also significant in the other disorder.

Hope I was clear, 
Lydia

Dear Lydia,

Thank you very much for your explanation！
We used SDM version 5.15, so these results were not corrected. We are going to try new version SDM-PSI.

Best,
You

Dear you:

I would like to know Where do I set the threshold value for multimodal analysis?

I can't find it in the interface of multimodal analysis. does the threshold is set directly during mltimodal analysis, or still need other steps to set the threhold after multimodal analysis?

Best
ZhangYi