 |
 |
Hi David, Thanks a lot for your answer! Well, the resolution goes to 1*1*1 when I register my functional scan to structural image. I will look into HCP-style more throughly to see if I can do my analysis in cifti graycordinate space. Thanks, Sara ________________________________ From: FSL - FMRIB's Software Library [[log in to unmask]] on behalf of David Van Essen [[log in to unmask]] Sent: Thursday, October 04, 2018 7:18 PM To: [log in to unmask] Subject: Re: [FSL] Dual regression taking so long! Sara, I noticed that you said "I am using an image size of 1*1*1" I presume that your fMRI voxels are much coarser, at least 2x2x2? If so, I don’t understand why you oversampled your functional data, given that it greatly elevates your processing time with no evident benefit. On a related note, have you considered doing your analysis “HCP-style” in cifti grayordinate space? It will make your computations even more compact and offer other benefits as well. See http://www.pnas.org/content/115/27/E6356 David On Oct 4, 2018, at 11:54 AM, [log in to unmask]<mailto:[log in to unmask]> <[log in to unmask]<mailto:[log in to unmask]>> wrote: Thanks a lot. With what we have right now, it should be approximately 2 months. It takes about 12 hours for one component out of 50, 5000 permutation to get done. Sara ________________________________ From: FSL - FMRIB's Software Library [[log in to unmask]<mailto:[log in to unmask]>] on behalf of Anderson M. Winkler [[log in to unmask]<mailto:[log in to unmask]>] Sent: Wednesday, October 03, 2018 5:51 PM To: [log in to unmask]<mailto:[log in to unmask]> Subject: Re: [FSL] Dual regression taking so long! Hi Sara, It's hard to tell without knowing the size of the data. You can get an estimate of when it will finish by considering the number of ICs, number of permutations, and how many have run already, then decide whether waiting that amount of time would be acceptable. All the best, Anderson On Mon, 1 Oct 2018 at 19:31, [log in to unmask]<mailto:[log in to unmask]> <[log in to unmask]<mailto:[log in to unmask]>> wrote: Hi Anderson, Thanks a lot for the explanation. I am not sure what you would call as "enough RAM" for such analysis. I use a a macOS with the following features: Processor: 4.2 GHz Intel Core i7 Memory: 16 GB 2400 MHz DDR4 Not sure if this is sufficient enough to run this analysis. Does it make a big difference if we upgrade our RAM to 64 GB? Thanks, Sara ________________________________ From: FSL - FMRIB's Software Library [[log in to unmask]<mailto:[log in to unmask]>] on behalf of Anderson M. Winkler [[log in to unmask]<mailto:[log in to unmask]>] Sent: Monday, October 01, 2018 3:44 PM To: [log in to unmask]<mailto:[log in to unmask]> Subject: Re: [FSL] Dual regression taking so long! Hi Sara, I would leave at 1mm and would be patient. Often it takes many months or years to collect data, at a high cost (time and money). My opinion is that it isn't worth to throw resolution away just to save a handful of extra days of processing which, btw, is being done by a computer automatically... Of course, this is a personal decision, but unless you don't have enough RAM (in which case running might be impossible without swap space, and even there, it would be way slower), it seems better to wait. Only make sure that the program is actually running, i.e., that you receive feedback in the screen (or in log files if that is the case), and that the process hasn't become a "zombie" (check with the command "htop" or "top"). All the best, Anderson On Mon, 1 Oct 2018 at 13:36, [log in to unmask]<mailto:[log in to unmask]> <[log in to unmask]<mailto:[log in to unmask]>> wrote: Thanks Anderson! I also noticed that I am using an image size of 1*1*1. I guess I have to make them smaller to make the process faster as well. Right? Thanks, Sara ________________________________ From: FSL - FMRIB's Software Library [[log in to unmask]<mailto:[log in to unmask]>] on behalf of Anderson M. Winkler [[log in to unmask]<mailto:[log in to unmask]>] Sent: Sunday, September 30, 2018 5:19 PM To: [log in to unmask]<mailto:[log in to unmask]> Subject: Re: [FSL] Dual regression taking so long! Hi Sara, Yes, it can be slow, particularly if the design includes regressors to accommodate repeated measures, many contrasts, spatial statistics, and if all ICs are tested. All the best, Anderson On Thu, 20 Sep 2018 at 12:59, [log in to unmask]<mailto:[log in to unmask]> <[log in to unmask]<mailto:[log in to unmask]>> wrote: Hi, I am running dual regression on 52 scans from 26 subject to find an intervention effect on their functional connectivity. We have 51 components from group level ICA. I let the command run yesterday morning, it finished the first two stages and created all the file for those stages by 7 pm. However, it's been about 23 hours that it has created only two files from stage 3. Here is the command that I am running: dual_regression groupmelodic.gica/melodic_IC.nii.gz 1 GroupICA.mat GroupICA.con 5000 /Users/sizadi/Desktop/Dual_Regression Post_006B_filtered_func_data_clean_standard.nii.gz Post_008B_filtered_func_data_clean_standard.nii.gz Post_012C_filtered_func_data_clean_standard.nii.gz Post_015C_filtered_func_data_clean_standard.nii.gz Post_027C_filtered_func_data_clean_standard.nii.gz Post_028B_filtered_func_data_clean_standard.nii.gz Post_032C_filtered_func_data_clean_standard.nii.gz Post_033B_filtered_func_data_clean_standard.nii.gz Post_034C_filtered_func_data_clean_standard.nii.gz Post_035C_filtered_func_data_clean_standard.nii.gz Post_039B_filtered_func_data_clean_standard.nii.gz Post_040B_filtered_func_data_clean_standard.nii.gz Post_042C_filtered_func_data_clean_standard.nii.gz Post_045B_filtered_func_data_clean_standard.nii.gz Post_047B_filtered_func_data_clean_standard.nii.gz Post_048C_filtered_func_data_clean_standard.nii.gz Post_050C_filtered_func_data_clean_standard.nii.gz Post_054B_filtered_func_data_clean_standard.nii.gz Post_058C_filtered_func_data_clean_standard.nii.gz Post_064C_filtered_func_data_clean_standard.nii.gz Post_066B_filtered_func_data_clean_standard.nii.gz Post_069C_filtered_func_data_clean_standard.nii.gz Post_072B_filtered_func_data_clean_standard.nii.gz Post_073C_filtered_func_data_clean_standard.nii.gz Post_075B_filtered_func_data_clean_standard.nii.gz Post_078B_filtered_func_data_clean_standard.nii.gz Pre_006A_filtered_func_data_clean_standard.nii.gz Pre_008A_filtered_func_data_clean_standard.nii.gz Pre_012B_filtered_func_data_clean_standard.nii.gz Pre_015B_filtered_func_data_clean_standard.nii.gz Pre_027B_filtered_func_data_clean_standard.nii.gz Pre_028A_filtered_func_data_clean_standard.nii.gz Pre_032B_filtered_func_data_clean_standard.nii.gz Pre_033A_filtered_func_data_clean_standard.nii.gz Pre_034B_filtered_func_data_clean_standard.nii.gz Pre_035B_filtered_func_data_clean_standard.nii.gz Pre_039A_filtered_func_data_clean_standard.nii.gz Pre_040A_filtered_func_data_clean_standard.nii.gz Pre_042B_filtered_func_data_clean_standard.nii.gz Pre_045A_filtered_func_data_clean_standard.nii.gz Pre_047A_filtered_func_data_clean_standard.nii.gz Pre_048B_filtered_func_data_clean_standard.nii.gz Pre_050B_filtered_func_data_clean_standard.nii.gz Pre_054A_filtered_func_data_clean_standard.nii.gz Pre_058B_filtered_func_data_clean_standard.nii.gz Pre_064B_filtered_func_data_clean_standard.nii.gz Pre_066A_filtered_func_data_clean_standard.nii.gz Pre_069B_filtered_func_data_clean_standard.nii.gz Pre_072A_filtered_func_data_clean_standard.nii.gz Pre_073B_filtered_func_data_clean_standard.nii.gz Pre_075A_filtered_func_data_clean_standard.nii.gz Pre_078A_filtered_func_data_clean_standard.nii.gz I was wondering if it's suppose to take that long. If so, it means that my analysis would be finish in more than a month. Please let me know what i should do to expedite the process. Thanks, Sara ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ######################################################################## To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1