 |
 |
Hi Ravi, Thanks for sending. The results are fine: for the first contrast there is nothing significant after FWER correction, and this is for both the simple t-statistic as well as for TFCE. For the second contrast there are significant results in the FWER corrected TFCE map, and applying FDR to it keeps (and extends slightly) these findings. It seems all good to me. Having said that, since the design is a paired t-test, when calling randomise, you need to use also the option -e, and supply a file with the exchangeability blocks (attached). All the best, Anderson On 30 December 2016 at 02:40, Bhatt, Ravi <[log in to unmask]> wrote: > Thank you! > > On Dec 29, 2016, at 7:45 PM, Anderson M. Winkler <[log in to unmask]> > wrote: > > Thanks Ravi. Will have a look and get back to you in a few days. > > All the best, > > Anderson > > > On 29 December 2016 at 00:45, Bhatt, Ravi <[log in to unmask]> > wrote: > >> Hey Anderson, >> >> Sorry for the late reply but I have uploaded the files to the link. I’ve >> put the proper mod_merg file being used, the mask (nifti file), the >> design.mat and the design.con files. >> >> Let me know if something doesn’t work or you need anything else. >> >> Thanks! >> Ravi >> >> On Dec 13, 2016, at 3:17 AM, Anderson M. Winkler <[log in to unmask]> >> wrote: >> >> Hi Ravi, >> >> Could you upload all your input files? You should have just received an >> automated email with instructions. Thanks. >> >> All the best, >> >> Anderson >> >> >> On 12 December 2016 at 20:59, Bhatt, Ravi <[log in to unmask]> >> wrote: >> >>> Hey Anderson, >>> >>> To run randomise the command I’m using is: >>> >>> randomise -i (*insert GM_mod_merg file here*) -o (*insert output file >>> name*) -d design.mat -t design.con -m (*GM_mask file*) -T -n 5000 >>> >>> To run fdr the command I’m using is: >>> >>> fdr -i (*insert uncorrected p-value file from randomise*) >>> --oneminusp -m (*mask file*) -q 0.05 —othresh=(*name of fdr corrected >>> file*) >>> >>> I’m comparing the files fdr corrected files labeled tfce_corrp_tstat1 >>> and tfce_corrp_tstat2 to the output files >>> >>> Thank you! >>> Ravi >>> >>> >>> On Nov 29, 2016, at 12:44 AM, Anderson M. Winkler < >>> [log in to unmask]> wrote: >>> >>> Hi Ravi, >>> >>> Could you show the command lines used to run randomise and to run fdr? >>> And which FWE files you are comparing with the FDR results? >>> >>> Thanks. >>> >>> All the best, >>> >>> Anderson >>> >>> >>> On 28 November 2016 at 03:30, Bhatt, Ravi <[log in to unmask]> >>> wrote: >>> >>>> Hello FSLExperts, >>>> >>>> I am currently running FSL-VBM on some of my data and wish to apply FDR >>>> correction on the uncorrected p-map images that is outputted as a result of >>>> the VBM scripts. There are plenty of results on the corrected p-value (FWE) >>>> image file, but when I apply FDR correction using the code on FDR page for >>>> randomise (below), nothing survives. This is strange to me because it was >>>> my understanding that FWE is more stringent that FDR. Any advice would be >>>> appreciated. Thank you! >>>> >>>> Ravi >>>> >>>> "fdr -i grot_vox_p_tstat1 --oneminusp -m mask -q 0.05 --othr >>>> esh=thresh_grot_vox_p_tstat1” >>>> >>>> >>>> >>>> ------------------------------ >>>> >>>> UCLA HEALTH SCIENCES IMPORTANT WARNING: This email (and any >>>> attachments) is only intended for the use of the person or entity to which >>>> it is addressed, and may contain information that is privileged and >>>> confidential. You, the recipient, are obligated to maintain it in a safe, >>>> secure and confidential manner. Unauthorized redisclosure or failure to >>>> maintain confidentiality may subject you to federal and state penalties. If >>>> you are not the intended recipient, please immediately notify us by return >>>> email, and delete this message from your computer. >>>> >>> >>> >>> >> >> > >