[Taxacom] Molecular data and synapomorphies

Tue Dec 2 02:33:38 CST 2008

I feel there may be a difference of opinion here.
Firstly I am not a taxonomist per se, therefore my comments were not 
intended to be viewed in the light of achieving a classification only.
Secondly, my comments were general and not applied to well studied 
groups like the primates.
Thirdly, I feel it is a little strong to use the phrase 'absolutely 
wrong' and then go on to claim that molecular data is 'suspected' to be 
in regulatory regions and the problems that may cause. The more 
enlightened molecular systematists are well aware of the type of data 
and sequences they are using (I do not class myself as a molecular 
systematist either, by the way - or indeed as anything much). Further, 
to state that selection may therefore strongly affect the analyses and 
interpretation of molecular data suggests that morphological trait data 
does not suffer from this problem. If this is indeed what was meant, I 
disagree. Of course, it is often possible to test for selection in any 
nucleotide sequences, and knowledge of the type of sequence you are 
analysing may be available - in better studies that go out of their way 
to obtain and use such information.
Finally, while molecular data may be based on a few exemplars, when a 
list of extensive analyses (cytology, biogeography et al) is given it 
should be noted that this list is missing for the vast majority of 
groups studied, and for some (ie cytology) it is very unreliable due to 
the lack of voucher specimens deposited and the low level of detail 
published.
I would re-iterate that a combined analysis of several data can be more 
informative than either a single type of data, or a priori measuring one 
data against another. In the group we have been working on 
(Antirrhineae) one can find morphological data to support different 
hypotheses depending on the characters chosen - including characters 
that follow the same patterns as the moleclar data. For classification, 
this means different authors will give different classifications based 
upon personal opinion, preference or even reasoned argument about the 
evolution of the characters themselves. Better, surely, to use alternate 
data sources for comparison for a more informed interpretation?
Alan

Richard Zander escribió:
> Been away and apparently missed a good discussion. My comments:
>
>  
>
> John Grehan:
>
> just went through the molecular data for 44 taxa and 1134 sequences
> which generated a parsimony analysis supporting a human-chimpanzee clade
> (Prasad et al 2008 Mol Biol Evol 25: 1795-1808) for sequences
> orthologous to a 1.0 Mb region of human chromosome 7.  In the coding
> sequences found no apomorphies at all for the large bodied hominoids so
> the result is cladistically impossible. 
>
> Comment: What happens in that if you have a terminal group of six
> exemplars, the ones with shared traits of any sort will pop out first in
> the cladogram and the ones without any additional shared apomorphies are
> tacked on terminally, even if they share nothing more than traits that
> put them in the groups. I've done what John did and found exemplars of
> different genera paired as sister groups on the basis of no shared data
> beyond being crowded together.
>
>  
>
> Alan DAvid Forrest:
>
> In this case molecular data are just another form of data. Incongruence 
> between data types requires analysis of what causes the incongruence, 
> not rejection of one data in favour of another based on a priori 
> preferences.
>
> Comment: Absolutely wrong. Molecular data have the failings of
> morphological data and few of the benefits. About a quarter of
> non-coding traits are suspected to be promoters and silencers and other
> regulatory features of the genome and therefore subject to selection,
> and therefore any analysis based on molecular analysis will have some
> convergence and parallelism. "Morphological" traits is a simplification
> for 250 years of analyzing the evolution (stabilizing selection that
> keeps species intact over millions of years), particularly finding
> conservative traits that organize other traits, using thousands of
> specimens and a multiplicity of techniques: morphometry, cultivation,
> cytology, mating behavior, biogeography, sequences, allozymes, cluster
> analysis including parsimony, coming up with basic taxonomic units and
> higher clusterings that minimize the chance of convergence and
> parallelism. Molecular data are based on a few exemplars and, if
> reliable, do give you an idea of genetic continuity and isolation
> events, but genetic continuity of a lineage and isolation events do not
> require speciation, particularly in the case of punctuated equilibrium.
> Combining analyses of evolutionary taxonomy (recognizing paraphyly and
> giving apposite rank to evolutionary novelties) and molecular analyses
> (genetic continuity) gives a good result that maximizes evolutionary
> information in a classification. 
>
>  
>
> *****************************
> Richard H. Zander 
> Voice: 314-577-0276
> Missouri Botanical Garden
> PO Box 299
> St. Louis, MO 63166-0299 USA
> richard.zander at mobot.org
> Web sites: http://www.mobot.org/plantscience/resbot/
> and http://www.mobot.org/plantscience/bfna/bfnamenu.htm
> *****************************
>
>  
>
> _______________________________________________
> Taxacom mailing list
> Taxacom at mailman.nhm.ku.edu
> http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
>
>   