[Taxacom] real molecular data

John Grehan jgrehan at sciencebuff.org
Fri Nov 14 06:57:41 CST 2008


Peter,

I think you misunderstand the purpose of my postings. It is not to
convince anyone of anything. Its to provide a gauge to my thinking as I
explore various avenues. I agree, the number of apomorphies per se is
neither here nor there (but the relative number is a different thing). I
have never seen much describing the actual data. I am curious. Perhaps
most are not. 

My principle concern is over the necessary cladistic validity of the
molecular data is the phenetic nature of alignment and the lack of
outgroup taxa to strongly corroborate putative apomorphies. This
convinces me that there are some cladistic problems with molecular data
that may explain why there may be a 'consistent' result in favor of a
human-chimpanzee clade that is cladistically erroneous. Maybe that will
not be convincing to anyone else. Maybe nothing will be. There is a big
industry behind molecular analysis. There is not for morphological
analysis.

Regarding the homoplasy (I presume you are referring to the gibbon
example I mentioned). Since there are so few outgroup taxa I took the
hard line - if the base was in the outgroup it was not a putative
apomorphy. If the molecular outgroup comprised all gibbons and all OW
monkeys (to be comparable to the scope of the morphological outgroup)
and it turned up in only one gibbon then one might argue that the gibbon
occurrence is an independent event. Until then....

John Grehan

-----Original Message-----
From: Hovenkamp, P. (Peter) [mailto:Hovenkamp at nhn.leidenuniv.nl] 
Sent: Friday, November 14, 2008 3:25 AM
To: John Grehan; TAXACOM
Subject: RE: [Taxacom] real molecular data

John,

If you ever want to convince others of your point of view, you can save
yourself the trouble of scanning thousands of positions until your eyes
start to twitch. The only result of this appears to be that you point
out how few apomorphies there are that are completely free of homoplasy.
Dismisssing characters when they show only a single instance of
homoplasy has been briefly in fashion in the late 70's-early 80's - it
was called compatibility analysis then, and was thoroughly discredited
under the name of clique analysis. 

Merits of the method aside - no-one is going to be convinced of your
point of view if that is all you can present as support for it. 


Peter Hovenkamp
Nationaal Herbarium Nederland, Leiden branch www.nationaalherbarium.nl  

> -----Original Message-----
> From: taxacom-bounces at mailman.nhm.ku.edu
> [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of John Grehan
> Sent: Thursday, November 13, 2008 6:39 PM
> To: TAXACOM
> Cc: Fetzner, Jim; Colin Groves
> Subject: [Taxacom] real molecular data
> 
> OK - just to prove I am nuts, I scanned every one of 21,507 sequence 
> positions for coding data in the Prasad et al 2008 study (Confirming 
> the phylogeny of mammals by use of large comparative sequence data 
> sets Mol.
> Biol. Evol. 25: 1795-1808). The coding sequences produced (by 
> parsimony
> analysis) a human-chimpanzee clade followed by gorilla and orangutan.
> 
>  
> 
> I found the following (although I admit in advance I may have missed 
> some. I had a twitch in one eye for a while)
> 
>  
> 
> Human-chimp - 2 apomorphic bases
> 
> Human gorilla - 1 apomorphic base
> 
> Human-African ape -  20 apomorphic bases
> 
> Chimp-gorilla - 1 apomorphic base
> 
>  
> 
> These apomorphies were defined as being absent from all other primates

> - in this case only 12 species. In morphology this limited sampling 
> would be usually recognized cladistically as an absurdly small 
> outgroup representation. The outgroup taxa in this case were:
> 
>  
> 
> 1 gibbon
> 
> 1 colobus monkey
> 
> 1 vervet monkey
> 
> 1 baboon
> 
> 1 macaque
> 
> 1 duski titi
> 
> 1 owl monkey
> 
> 1 squirrel monkey
> 
> 1 marmoset
> 
> 1 galago
> 
> 1 mouse lemur
> 
> 1 lemur.
> 
>  
> 
> The best I think that may be said of the molecular evidence supporting

> the human-chimpanzee relationship is that it is 'overwhelming' if one 
> ignores the non-cladistic nature of alignment, and the small outgroup 
> sampling that create apomorphies as an artifact of small outgroup 
> sample size. For example, at site 14,466 there humans and African apes

> share A and that would have been an apomorphy if it were not for the 
> inclusion of that one gibbon where it also turned up.
> 
>  
> 
> So my conclusion is that the molecular claims of proving the 
> human-chimpanzee relationship is more propaganda than science, and 
> more phenetics than cladistics.
> 
>  
> 
> I rest my case (for now).
> 
>  
> 
> John Grehan
> 
>  
> 
> Dr. John R. Grehan
> 
> Director of Science
> 
> Buffalo Museum of Science1020 Humboldt Parkway
> 
> Buffalo, NY 14211-1193
> 
> email: jgrehan at sciencebuff.org
> 
> Phone: (716) 896-5200 ext 372
> 
>  
> 
> Panbiogeography
> 
> http://www.sciencebuff.org/biogeography_and_evolutionary_biology.php
> 
> Ghost moth research
> 
> http://www.sciencebuff.org/systematics_and_evolution_of_hepialdiae.php
> 
> Human evolution and the great apes
> 
> http://www.sciencebuff.org/human_origin_and_the_great_apes.php
> 
>  
> 
>  
> 
> _______________________________________________
> Taxacom mailing list
> Taxacom at mailman.nhm.ku.edu
> http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
> 




More information about the Taxacom mailing list