On Fri, 23 Jun 1995, Burton Guttman wrote:

>
> I really wanted to take this topic off Tweeters, because it seems to be
> one of those technical questions of interest to very few people, which

Hey, but I'm one of the few!! Eggheads speak up and back me up here! :)


> If this guy uses PCR (polymerase chain reaction), he might have some
> primers that would select out one particular sequence to study, one that
> he knows (but how does he know?) is specific to the animal and not found
> at all in the commensals. But that would only give him at most a thousand
> or a few thousand base pairs compared with the enormous amount of DNA in a
> mammalian genome. That seems like a terribly small sample of the genome
> on which to base any conclusions about taxonomy or phylogeny.


A thousand or few thousand base pairs is potentially a HUGE wealth of
information. Nobody looks (or could ever look) at the whole genome.
It's real big. So you pick a piece to look at. Some pieces are more
informative than others, but a piece 50 base pairs long, if variable
enough in the gene pool, could potentially tell you what subpopulation an
individual came from. With a thousand base pairs worth of fragments,
chosen carefully, you can tell lots.

> For such a
> study, isn't it a whole lot easier and more reliable to take a small blood
> sample?

Well, the reason people are working on the DNA from poop (tm) technique is
that it is easier (and less invasive) than taking blood samples. It's
not easy to take blood from free living wild animals. I've been doing just
that with Song Sparrows for the last three weeks, and I speak from
experience. If the technical details are worked out, fecal sampling could
probably be just as reliable as blood for most (but not all) purposes.

Chris Hill
Seattle, WA
cehill at u.washington.edu