Tagged: Subgraph analysis
May 1, 2018 at 1:15 pm #21926
Hello BRAPH team,
I can’t seem to run subgraph analyses on my data and I am unsure why. I set up a weighted undirected graph with pearson correlations and I am including negative correlations (however, I have also tried setting negatives to zero and I get the same error described below when I try to run subgraph analysis). Just in case it’s relevant, I entered my subject-level data as correlation matrices (as described in another forum post) rather than timeseries data.
I set my community structure based on one group of participants by entering the community structure window, selecting Newman algorithm, gamma = 1, and clicking “Calculate”; then I checked “Fixed structure”, clicked “Set”, and closed the community structure window. I next opened the Subgraph analysis window, selected the first module (module1), clicked “set” and then “start analysis”, which returned the following error in matlab:
Matrix dimensions must agree.
Error in GUIfMRIGraphAnalysis/update_matrix (line 714)
A = A+ga.getA(indices(i));
Error in GUIfMRIGraphAnalysis/setup (line 1204)
Error in GUIfMRIGraphAnalysis (line 1188)
Error in GUIfMRIGraphAnalysisSubgraph/cb_menu_ga (line 575)
Error using GUIfMRIGraphAnalysis/cb_calc_sub (line 483)
Error while evaluating UIControl Callback
Can anyone provide some guidance on what’s going on here and how I can fix it?
Thank you very much!
JeffJune 1, 2018 at 1:45 pm #22126
First, I would like to apologize for getting back to you late. I checked Braph for your problem and I was able to perform subgraph analysis, which means that there are not some obvious problems in the code.
Therefore, if you have not been able to solve the problem, I would like to ask you if you could send me an example of the matrices that you tried to analyze. In that case I could try the analysis with them and locate the source of the issue.
June 29, 2018 at 7:41 pm #22275
- This reply was modified 1 month, 2 weeks ago by Mite Mijalkov.
Thank you for your response.
I still haven’t had any luck with subgraph analyses, but I thought of something that might be the issue. Based on directions you provided in another post, I’m directly entering connectivity matrices into Braph, rather than timeseries data. Could that be the problem?
I can send you either a .fc file or a .fga file–do you have a preference? Also, what’s the best way to send them? I might be overlooking something but I don’t see any way to send an attachment directly through the forum.
Thanks a lot,
JeffJuly 9, 2018 at 11:39 am #22317
I do not think that directly entering matrices instead of calculating them within Braph should be the problem due to the organization of the software. However, as you said, it would be better if I have the sample file in order to check it directly.
Please send me the .fc file and let me know which correlation option you are using to calculate the matrices. You could send them to me directly on my e-mail: firstname.lastname@example.org
Looking forward to hearing from you.
- This reply was modified 1 week, 4 days ago by Mite Mijalkov.