HSP70 evolution

aroger at ac.dal.ca aroger at ac.dal.ca
Sun Jan 29 18:30:30 EST 1995


In article <3fjvs4$lq9 at rebecca.albany.edu>, labonnes at csc.albany.edu (S. LaBonne) writes:
>>
> 
> You've put your finger directly on the thing I find most mystifying
> about some of these debates: that the pros in molecular phylogeny,
> whom one would expect to understand these things, seem repeatedly to
> lose sight of the basic distinction between gene trees and organism
> trees- a distinction familiar even to relatively untutored outsiders
> like me.  


The point of the discussion is obvious- where does the root of the
hsp70 tree fall?  

Certainly one issue surrounding the hsp70 topology is whether this
is a gene tree tracking orgnanismal phylogeny or whether there
have been gene-transfer events which have given rise to "peculiar"
affinities of some of the taxa in these trees (e.g.- archaebacteria
and gram positives being specifically related).  Who are you 
accusing of forgetting about this?  Larry Moran discussed at 
length the alternatives of lateral transfer accounting for features
of the tree versus Gupta and Golding's fusion hypothesis for the
origin of eukaryotes (in the latter view it is held that the gene phylogeny of 
hsp70 is accurately indicating that one of the ancestor's of
eukaryotes was a gram negative eubacterium).

Simply saying-- "its simple; gene phylogeny doesn't always track
organismal phylogeny" -- is entirely MISSING THE POINT.  The central
problem is;" is hsp70 tracking  organismal phylogeny?" -- and currently the
molecular data (perhaps the only very strong data bearing on the
phylogenetic question at hand) are giving a mixed message.  We 
cannot easily decide whether the hsp70 tree shows a different
topology than the elongation factors or rRNA because of lateral 
transfer events of the former or the latter or whether there was
a fusion event.  If lateral transfer does explain hsp70 then 
between which cells did it occur?  Why do several other genes
show similar topologies to hsp70? 

Andrew Roger
aroger at ac.dal.ca



More information about the Mol-evol mailing list