On 12 May 2011 18:25, Cinly Ooi <[log in to unmask]> wrote:

Dear All,

Thank you for continuing to provide quick (and accurate) responses to all my questions.

I am in the process of finalizing the feat scripts for a project and needed the cope*.nii.gz and varcope*.nii.gz in standard space (AAL template space, i.e. 91x109x91...)  to extract statistics value for individual AAL region.

Reading  http://www.fmrib.ox.ac.uk/fsl/feat5/output.html says I can get them in standard space in reg_standard directory

An old  posting (2007) found in http://www.biac.duke.edu/forums/topic.asp?TOPIC_ID=1062, says reg_standard is created by second level. I tried it and it works. I chosen the stats to be single group average. I was wondering whether is there an option in first level analysis that can do it.

My question are:

(1) Is it possible to generate reg_standard in 'First Level Analysis' easily?

(2) To create all the standardspace images quickly, I plan to run as many second level analysis as possible. Ideally, this means one per subject.

I do not have a SGE machine setup unfortunately

However, 'Higher Level' analysis UI requires me to put in two datasets, regardless of whether is it 'Lower level FEAT directory' or 'cope files'. As I only need reg_standard directories inside individual subject's directory, can I leave the second FEAT directory/cope file blank?

I tried it. As expected, I got failure at 'Higher Level Input Files Preparation', but the reg_standard/stats/cope*.nii.gz and varcope*.nii.gz were generated for the first FEAT directory. Can someone confirm that the cope and varcope files are good, i.e. I can use it to extract statistics? What I am worry about is that these are intermediate results and a process will convert them to the final results later. However, looking at how FEAT uses temporary files to hold intermediate data, I do not think they are temporary results and therefore  they are OK. Am I correct?

Many thanks in advance and hope to hear from you all soon.

Best Regards,
Cinly