>It has been done, the retroviruses that carry oncogenes are examples of
>this.
Yes, I know retroviruses carry oncogenes, and also that retroviruses
can carry foreign genes into the genome of an infected host.
But that's not the same thing as the hypothetical experiment I sketched,
which, to my knowledge, has not been done as I described it.
I'm talking about a hands-off experiment where we dump two eukaryotic
species and some retroviral vectors together -- NOT where a retroviral
vector is carefully engineered by a molecular biologist to carry a cloned
gene into a foreign host. To demonstrate, in real time, that
Gene A from species X ----> moved via viral vector Q ----> into species Y
shouldn't be hard to do, right? Just put everything together (X, Q, Y)
in a culture and let the retroviruses do their stuff. If it happens in nature in
evolutionary history, it ought to happen pretty easily in a lab, with no
help from molecular biologists. It's THAT experiment that I'd like to see.
about lateral gene transfer in the evolutionary past are typically driven by
"heck, it might have happened" speculation -- because there happen to be
such things as retroviruses. This becomes especially problematic when
anomalous gene distributions (e.g., as mentioned by Joel Duff) are being
explained away, in the absence of any knowledge of the transfering vector.
Paul Nelson
University of Chicago