So much hor-gene-xfr in Prokarya that whole-cell family-trees

RobertMaas at YahooGroups.Com RobertMaas at YahooGroups.Com
Sat Dec 28 00:54:16 EST 2002


I tried on 2002.Dec.24 to post this to bionet.molbio.evolution,talk.origins
but it never appeared, so I'm trying again:

More than four years ago, I began advocating generating a complete
"tree of life", not for whole-organism/cell lines, but for each gene
separately, thereby removing the bias that assumes no significant
amount of horizontal gene transfer ever happens, and thereby be able to
directly answer the question how much horizontal gene transfer has
happened, as well as resolve other questions about intra-cellular
symbiosis including the origins of the various kinds of Eukaryotes.
For example, see what I posted:

1998.Nov:
http://groups.google.com/groups?selm=73eple%24ii3%245%40supernews.com

1998.Dec:
http://groups.google.com/groups?selm=74d95v%24ivu%2411%40supernews.com
http://groups.google.com/groups?selm=74lejb%2485f%244%40remarQ.com

1999.Jan:
http://groups.google.com/groups?selm=7817ko%241lg%2411%40remarQ.com

1999.Feb:
http://groups.google.com/groups?selm=7bc4mm%24knp%40net.bio.net

Then this past evening (2002.Dec.23), while I was at the library
catching up with recent back issues of SCIENCE, I was pleased to find
in the issue for 2002.Nov.22, on page 1538-39, an article reporting
exactly that kind of research accomplished the Summer of 2001.

One genome each from five groups of bacteria were compared and found to
have about 200 genes in common (presumably *not* identical, but rather
similar enough that the researchers believe each of the 200
matches-of-five evolved from a common ancestor more recent than any
common ancestor of "different" genes).

The researchers then studied these 200 shared genes and found that
about 50 of those common genes are for photosynthesis (quite a
complicated process obviously). For each of these 200 groups-of-five, a
family tree was constructed. Then these trees were compared per which
of the five groups of present-day organisms they currently reside in.

Result: Of the fifteen possible such images of family trees, all
fifteen occurred. This shows that over the time scale since ancestors
of these five cell lines diverged from a common ancestor (*), the genes
basically evolved independently, rather than co-evolving within
whole-cell (whole-genome) family lines, at least to the fifteen degrees
of freedom which can be measured among only five genomes. This means
that massive amounts of "horizontal transfer" must have occurred after
ancestors of these five strains of bacteria diverged from their
presumed very ancient common ancestor.

* (Because of these results, it may actually be meaningless to speak of
whole-cell family tree over such long times. Over much shorter time
spans, perhaps mostly co-evolution within cell lines occurs, and it's
meaningful to draw whole-cell-line family trees, and if we artificially
exclude photosynthetic genes and any others that experienced horizontal
transfer, we might be able to derive whole-cell-line family trees going
back all that time, but over the long time span and including the
photosynthetic genes, each recent whole-cell-line is really like a
Chinese-restaurant menu that mixes several separately-evolving cell
lines of earlier times to make each modern cell-line-tree. Possibly not
just these 50 genes but virtually *all* genes in Prokarya have
experienced multiple episodes of horozontal gene transfer, and if we
artificially exclude all these genes there's nothing left on which to
base a whole-cell-line (whole-genome-minus-some) family tree. I
encourage further per-gene family-tree analysis, combined with mapping
these trees back into the list of cells studied, to resolve this
question of measuring the relative amounts of co-evolution vs. Chinese
restaurant mix-and-match.)




More information about the Mol-evol mailing list