01 February 2022 0 276 Report

I'm attempting to do conjunction analysis in SPM8 with PET data from subjects before and after treatment, to analyze the correlation between PET imaging results and gene expression

I first created a multiple regression batch design with all the scans, before and after treatment for all subjects, and entered a few covariates (genes and their expression level in each subject before and after). I then created separate t-contrast weight vectors for each covariate, thus generating about 10 contrasts. Then in results > SPM.mat file > contrast manager, I selected multiple contrasts at once, and then clicked "Done". I was then prompted in the results window; I selected conj, no masking, p value adjustment to 0.63 (a value that I settled on through trial and error that generates a reasonable number of significant voxels), and a voxel threshold of 30. When I use a P value of 0.01, no voxels survive.

Now I'm quite curious as to what these results mean, and how I should interpret it. From what I've gleaned online, I understand that I can only say I found activation in at least of the scans (the contrasts that I entered into conjunction analysis), activation being a significant result from the spm contrast. Also am I even doing conjunction analysis correctly? Thanks

Similar questions and discussions