chloroplast and other genes (was Lucy in Newsweek)

[B.J.Tindall]

I am not sure whether anyone said "reject ALL molecular data as phenetic
and unreliable". Its just a case of how we tend to interpret the sequence
data. We tend to accept that when two genes share 90% sequence similarity
they are "closer" together than the third gene which shares 80% sequence
similarity with the other two. By applying "parsimony of logic" we would
tend to exclude the fact that the higher degree of similarity arose due to
effects such as parallelism, gene transfer, or perhaps even other effects
which we consider not be the "logical" - thus "overall similarity" (i.e.
phenetic influence) plays a subtle underlying role. Cladistic analysis
would then build on that concept. This might also influence what Don
Colless calls "gut feeling". Certainly my earlier e-mail which includes
reference to a "phenetic" approach was not meant to imply that this makes
molecular data unreliable AT ALL. The "unreliability" reputation of a
phenetic approach has come via the misunderstanding that phenetic =
phenotypic. In the early days of the 16S rRNA cataloging the opinion was
expressed that simple Sab values (i.e. measures of overall similatiry) were
phenetic, to which the counter argument was that there were "phylogenetic"
(but were they using cladistic analysis!?). Under certain circumstances
phenetic approaches will also reflect evolution accurately and my
impression is that some of the discrepancies bewteen molecular and
moprhological data arise simply because one accepts that a "greater degree
of genetic similarity" reflects a closer degree of evolutionary relatedness.
My impression is that some 10 years ago we were given the impression that
once complete genomes were available we would have all the answers. This
reminds me of approaches in the early 1960s, where the goal was to obtain a
sufficiently broad spectrum of data which would give a "relibale result".
Even in the case of sequence data we are now moving into a period of
selecting the "correct" data set and weighting, which seems to me to be
something we went through with the phenotypic data some decades ago.
Brian


At 22:41 1.4.2004 -0600, Ken Kinman wrote:
>     As for ape phylogeny, I still don't know enough to hazard a guess
whether nuclear genes are generally more reliable than mitochondrial genes.
 I can only say that individual bases changes are generally the least
reliable, and I am more comfortable relying on strings of bases that are
inserted, deleted, or modified in other less probable ways (than single
base changes).  There is also the higher order arrangement of whole genes
that genome sequencing will uncover.  Those who reject ALL molecular data
as phenetic and unreliable are failing to appreciate the potential of this
broader genetic "morphology", although morphologists are right to question
phylogenies based only on the less reliable patterns of individual base
substitutions.   None of this is cut and dry, and certainly not simple.  We
definitely have alot to learn about weighting characters, both molecular
and morphological, and neither approach has all the answers.
>          ------- Cheers,
>                         Ken Kinman
>


********************************************************************
* Dr.B.J.Tindall      E-MAIL bti at dsmz.de                           *
* DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH *
* Mascheroder Weg 1b, D-38124 Braunschweig, Germany                *
* Tel.: ++ 531 2616 0 (general)                                    *
* Tel.: ++ 531 2616 224 (direct)                                   *
* Fax:  ++ 531 2616 418                                            *
*                                                                  *
* Homepage: http://www.dsmz.de/index.html                          *
* E-MAIL: contact at dsmz.de (general enquiries)                      *
*         sales at dsmz.de (sales)                                    *
********************************************************************