Open course ware

ISAAC SAMUEL KOHANE: So let me give you a case history. This is work that I've done with one of my former interns, actually, who's a superb scientist at the Dana-Farber Cancer Institute, David Rowitch, who also is a part time neonatologist. I call this finding the needle in the haystack. And it's about studying the cerebellum. So I impressed, I hope, upon you on the last lecture, just how bad microarrays are. How noisy they are, and how difficult it is to interpret them. So now, I'll show you how-- darn it-- how, if you actually do the computation right, you can actually extract useful biomedical information out of it. So the cerebellum is a nice system to study because it has this nicely layered structure. The cerebellum is this part of the brain that sits in the posterior fossa, at the back. And has this very nicely layered structure which is detailed in this cross section here so that each layer has this distinct group of cells. And often, these group o...

ISAAC SAMUEL KOHANE: Also, I forgot to mention at this point, the output of microarray studies is foreign to basic biology researchers. They're used to looking at three or four or five or 20 numbers and performing some easy analysis in an Excel spreadsheet. But the point-- or doing a BLAST of one gene at a time. But the data from these microarrays does not easily load into standard tools. The analysis is non-standard. Excel does not cut it. Excel can barely load some of these data sets. And so all this is actually representing an interesting turn. So this is the other point of view, which is now that the standard biologist tools are no longer fitting nicely onto the desktop productivity suite that they previously had. And this is, in some way, a full circle. It used to be, of course, that computers were run by the high priests in the computer center. And not only that were they-- did they know about how to boot-up the computer and to make the computer do whatever it d...