Hello! I just finished computing a single-subject GLM with 3dDeconvolve and now I have a stats bucket file with Regression Coefficients and t maps that look accurate (YEA!). Now, I just want to threshold (and perhaps cluster) my output t maps so that I can make a publishable figure. I am indifferent as to whether I use family-wise error or the FDR correction- it won't matter in my case because my t values are really large. I'm seeking an automated method to properly apply a multiple comparison correction to the t maps in the bucket file without any human interaction (no GUI, no button pushing). Assuming that I want to threshold the 3rd image in the bucket file "stats.nii.gz" at P<0.01, what would be the command or commands to do this?
For example, I have been told by a colleague that a good multiple correction method may be to just threshold the images on the basis of the q value. In the GUI, when I select the "Coef" as the "Olay" and the "Tstat" as the "Thr", I can adjust the intensity threshold to 3.909 until I get p=9.9^-5 and q=0.0091. Is 3.909 my two-tailed p< 0.01 FDR corrected t map threshold? Can I obtain the value of the intensity threshold (3.909) with an AFNI program so that I don't need to open the GUI? If anyone has scripted the thresholding process, I would love to see your code. Thank you!
