In article <badger.844017296 at phylo.life.uiuc.edu>,
Jonathan Badger <badger at phylo.life.uiuc.edu> wrote:
>lamoran at gpu.utcc.utoronto.ca (L.A. Moran) writes:
>>>You can plug anything you want into "gene X". All such statements have about
>>the same scientific legitimacy. Why do some people feel that rRNA genes
>>are special in this sense? This is an argument that seems to be widely
>>disseminated but rarely questioned. Can anyone supply a reference to an
>>article makes the case that rRNA genes cannot be horizontally transferred
>>but others can?
>Of course all genes be horizontally transfered, in the sense that it
>would be physically possible. However, the more interesting question
>is "is the copy we see today in the genome derived from a horizontal
>transfer event in the past".
>Let's consider a prokaryote hundreds of millions years ago that has
>just recieved a gene by horizontal transfer. If the horizontally
>transfered gene doesn't confer any selective advantage, then why would
>the particular organism receiving the copy out of its billions of
>sisters be the ancestor of a phylogenetic group we are studing today?
>We only see the descendants of the winners of evolution.
I am arguing that there is no particular reason to suppose that rRNA genes
are *less* likely to be transferred horizontally. Many of the proponents of
the Three Domain Hypothesis dismiss the conflicting dendrograms by
postulating that horizontal gene transfer has occurred. Of course this
tactic could also be used on their favorite genes (ie. rRNA genes) except
that they claim otherwise. Their logic doesn't stand close scrutiny.
You are asking a different question. You are making the assumption that
all (most?) of molecular evolution is due to natural selection therefore
you wonder why any horizontally transferred gene would become fixed in
the population to begin with. Your assumption is not correct. We do not
see only "winners of evolution" in the sense that all organisms had to
have a selective advantage over their relatives. Much of evolution at the
molecular level is due to genetic drift and many of the mutations (alleles)
are neutral or nearly so. There is a significant probability that an
individual that acquired a gene by lateral transfer could give rise to
more offspring by chance. The transferred gene could eventually become
fixed in the population even though it did not confer selective advantage.
Most of the gene dendrograms that you see in the journals are due to the
fixation of neutral alleles by genetic drift. That's why we can talk about
an approximate molecular clock. (If most of the changes were due to the
fixation of alleles by natural selection there would be no obvious reason
to presume that this occurs at a steady rate.)
>So, unless you believe a horizontal transfer of rRNA would confer a
>great selective advantage, the chances of rRNA genes today being
>derived from a horizontal transfer event in the past are very slim
>indeed.
I don't agree with the logic behind your argument. The horizontal transfer
of an rRNA gene may not confer any selective advantage but nevertheless the
transferred gene could become fixed in the population. This will, of course,
be a rare event but it is quite possible. If the transferred gene confers
a selective *disadvantage* then purifying selection will likely eliminate
that individual or its offspring.
Are you assuming that all of the proposed examples of lateral transfer are
due to selection because the transferred gene conferred greater fitness?
>Phylogeny is about evolution. Wherever thinking about it, one needs to
>think evolutionarily.
I agree. But not everyone does! (-:
Larry Moran
