in reply to Re: Deducing Ideal Groups
in thread Deducing Ideal Groups

Interesting. I've been stirring this problem around for a couple of months, bringing it to the forefront occasionally. In the past, I looked at a package called AutoClass, from NASA. It rang a bell right away when I saw your microarray picture, as that's exactly what's on the front of the AutoClass page. However, I've not been able to make the jump from items with many varing-but-enumerated characteristics to items with two states.

Thanks for the info, I'll study it more carefully today. BTW, my workstation is Debian, and the answers will ultimately be used by an application that runs on IBM mainframes (z/OS).