***** To join INSNA, visit http://www.insna.org ***** Dear Jay, I'd recommend DEDICOM algorithm for pattern analysis in directed graph. You may have a look at this answer http://datascience.stackexchange.com/questions/6834/cluster-directed-graph-into-dag Cheers, Kasra On Sun, Jan 8, 2017 at 10:49 PM, Hegde, Jay <[log in to unmask]> wrote: > ***** To join INSNA, visit http://www.insna.org ***** > > Hi Everyone, > > > > We have two NxN connectivity matrices that define *weighted*, > *directional* interactions among the *same set of N nodes* under two > different conditions, A and B. Since the interactions are directional, our > connectivity matrices, while being square, are *asymmetric*. > > > > We want to identify the interactions that are significantly different > between the two conditions. That is, how do the nodes talk to each other > under condition A that do not do under condition B? Of course, we want to > do more than just calculating the difference between the two matrices. For > one thing, we know that if the matrices are symmetrical, there are toolkits > for comparing them. > > > > But we’re not sure how to compare our asymmetric matrices. Any ideas about > (1) how to do it, and (2) any toolkits that implement the solution? > > > > Thank you very much, > > Jay > > > > Jay Hegdé PhD > > Medical College of Georgia > > Augusta University (formerly Georgia Regents University) > > Augusta, GA 30912-2697 > > > > > > > _____________________________________________________________________ > SOCNET is a service of INSNA, the professional association for social > network researchers (http://www.insna.org). To unsubscribe, send an email > message to [log in to unmask] containing the line UNSUBSCRIBE SOCNET > in the body of the message. _____________________________________________________________________ SOCNET is a service of INSNA, the professional association for social network researchers (http://www.insna.org). To unsubscribe, send an email message to [log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of the message.