In article <timi-230195180310 at kos4mac15.berkeley.edu>, timi at mendel.berkeley.edu (Tim Ikeda) writes:
> About the comparisons to the gram positives...
> Are there any HSP70 sequences from the high-GC gram positives or are
> they all from the low-GC, (Bacillus, Clostridium, Mycoplasma) branches?
> It's just that the low-GC's seem to have really been on the move,
> evolutionarily speaking. They're quite divergent. I just wonder how
> the comparison goes when one includes a representative sampling from
> the "Actinomycetes" (high-GC) subgroup of the gram-positives, which
> appear a bit more like the gram-negatives in many ways.
>> Regards, Tim Ikeda (timi at mendel.berkeley.edu)
With regards to the hsp70 dataset, there are both actinomyctes
(high G+C) and low G+C gram positives included. Curiously,
in the recent paper by Gupta and Singh (Current Biology 1994,
Vol 4, #12), the Halobacterial sequences cluster with some of
the high G+C gram +ves (Streptomycetes and Mycobacteria) whereas
Thermoplasma and Methanosarcina sometimes cluster together
specifically related to the Clostridia (a subgroup of the low
G+C gram +ves). One question you could ask is: are the
archaebacteria appearing polyphyletically derived from the gram
positives because of G+C coding sequence bias? It might be
interesting to see if coding sequences for hsp70 have become
biased in amino acid sequence as a result of G+C pressure (after
all, it is rather odd that Halobacteria which are notoriouslyGC-rich,
should happen to cluster with high GC gram+ves while the low GC
archaebacteria whould show affinity to low GC gram +ves). Gupta
and Singh argue that this is not the case.
Andrew Roger
Dept. of Biochemistry
Dalhousie University
Halifax
aroger at ac.dal.ca
