Take the 2-minute tour ×
Biology Stack Exchange is a question and answer site for biology researchers, academics, and students. It's 100% free, no registration required.

We've begun to try out the SCA Matlab toolbox (latest version) downloaded from Dr. Rama Ranganathan's website, and, following the included tutorials, would like to apply it to our protein family. The SCA calculation involves determining independent components (IC) of positional covariation by applying the "Infomax" algorithm to a few top eigenmodes (that have been obtained from the positional correlation matrix). Here, we observed that the output (IC vector weights) seems to depend on the no. of top eigenmodes chosen as input (i.e. on choice of the parameter kmax). If this is indeed the case, then is there some "working rule" to optimally choose kmax (say, based on the number of significant eigenmodes turning up)? Since this isn't made very clear in the tutorials (at least to me), I thought of trying my luck here.

Would really appreciate some guidance/pointers from SCA users. Thanks a lot!

share|improve this question

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.