NHE Mar 24 Notes
Doing supervised clustering with a down regulated gene. (ITPA - from the 6 highly differentiated genes)
Setting the threshold value to 0.0256 produces correct clustering vs 0.0257 which has improper clustering. The gene that makes the difference is INTS7
Also did supervised clustering around LGR5 (Campbell recommended)
Re-did clustering generally with mean range from 1 to 10000 and qvals = 0.01
Found a list of 44 genes, but not all the same. 6 do not overlap. Don't know why. Have that list.
Doing KAAS for ITPA (down regulated) and 44 genes found with the larger range
want to do more specific data mining with the GO IDs and KEGG Pathways