Update Consensus network code

[alexanderlewis99]

Dr. Califano suggests that ARACNe has a bunch of false negatives. Therefore if a gene comes up more frequently it is more valuable. Therefore, perhaps we should modify consensus network so that it cuts values of a gene that is only in one network.

[alexanderlewis99]

ARACNe3 has an algorithm to create consensus networks. Perhaps consult with Andrew to ensure our consensus approach is similar/same