I have 1200 TFs. I should make a 1200 in 1200 matrices to compare them and get TF-TF interaction. How can I do it? What programming can I use? yes, I am trying to do is look for evidence of interaction (Y/N) between 1200 transcription factors by making a 1200 x 1200 matrix to look pairwise interactions between each of the 1200 TF. My data are human protein and were downloaded as file text. But I don’t know how to do it.
3 answers
Use a protein-protein interaction databases and use one of those to build your matrix, See other biostar threads for more info on PPI database. e.g. What Is Your Experience With The String (Interactions) Database?
Pairwise interaction is not always a good approach to compute/establish interaction between such a huge list of TF's. I would suggest you to read the following article, " Scoring overlapping and adjacent signals from genome wide ChIP and DamID assays". This reviews gives an insight to the statistical methods which could be applied to study the interaction pattern between TF's. furthermore, you may also use Cooccur package from David huen.
Log in to answer this question.
whar is TF ? a transcription factor ? What kind of data do you have ? What do you mean by "comparing" two TFs ?
and what would the data (numbers) you put in your 1200 x 1200 matrix represent?
I believe what shahram is trying to do is look for evidence of interaction (Y/N) between 1200 transcription factors by making a 1200 x 1200 matrix to look pairwise interactions between each of the 1200 TF in question. What I would suggest is a protein-protein interaction database but I dont know how accurate these are.