Brain Connectivity Toolbox requires 'self-self' connections along the main diagonal to be zero

[Chris Whelan]

Hi all,

I was wondering whether anyone here has used connectivity matrices generated in ExploreDTI in software pacakges like the Brain Connectivity Toolbox (BCT)? 

Crucially, BCT requires 'self-self' connections along the main diagonal to be zero. I noticed that ExploreDTI does not set these values to zero. BCT also has a hard time recognizing 'NaN' values in the connectivity matrices, so most of these need to be converted to zero.

How has anyone on this message board got on with BCT?

Thanks,

Chris 

[Mohammad Hadi Aarabi]

Dear Chris,

You can use CM(isnan(CM))=0 to solve your problem. 

Best,

Mohammad

[Mohammad Hadi Aarabi]

Also you can use this code to save your mat file to txt file :

CM(isnan(CM))=0; fid = fopen('subject01.txt','wt'); for i=1:size(CM,1) fprintf(fid,'%12.6f\t' , CM(i,:)); if i < 90 fprintf(fid, '\n'); end end fclose(fid);

[Chris Whelan]

Thank you, Mohammad! Do you have any idea how the main diagonal can also be set to zero? This appears to be a requirement of the Brain Connectivity Toolbox. 

Best,

Chris 

[Shaofeng Duan]

CM(isnan(CM)) = 0;
CM_temp = triu(CM,1); % 'doc triu' for help
CM = CM_temp + CM_temp';

在 2014年12月17日星期三UTC+8上午5时03分59秒，Chris Whelan写道：