[Taxacom] Fish phylogenetics paper

[John Grehan]

I found this to be a very nice summary of my understanding about the
limitations of molecular characters and why they do not have any
necessary precedence over morphology (which I consider more genetic than
DNA molecules). Work on primate phylogeny for humans and orangutans has
been globally condemned (and on this list) as preposterous simply
because it is based on epigenetic (if I am using that term correctly)
rather than DNA molecular characters, and that it does not subordinate
the epigenetic finding to that of molecular similarity. The constraints
outlined below show why molecular comparisons, apart from being
metaphysical in that the characters and character states are defined by
an algorithm rather than direct observation in nature, may represent
overall similarity than necessary relationship. 

That is not to say that I would be opposed to molecular phylogenies
(after all they definitely produce useful biogeographic information),
but when there is a conflict with morphology there is a problem that
cannot be solved simply by assuming the molecular comparison is better.
In some cases the morphological work is problematic also, but that can
be assessed in a way that the molecules cannot.

I am less concerned with the algorithms used to cluster or relate
characters - I am not a mathematical theorist so I have no necessary
prejudice. I use parsimony methods simply because many use it and it
does not become a side issue for the result. Other clustering techniques
might do better with the same data and that is ok with me.

John Grehan

-----Original Message-----
From: taxacom-bounces at mailman.nhm.ku.edu
[mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Bob Mesibov
Sent: Wednesday, May 12, 2010 7:07 PM
To: kmagnacca at wesleyan.edu
Cc: TAXACOM
Subject: Re: [Taxacom] Fish phylogenetics paper

Hi, Karl.

So you didn't find the discussion on characters and homology
interesting? I did, because Mooi and Gill seem to be asking molecular
systematists to be aware of, and think about, character homologies in
the same way that morphological systematists do. And it's pretty obvious
that many molecular systematists couldn't care less about molecular
characters, the exceptions being those working on the higher levels of
sequence data, i.e. whole genes, gene families, etc.

I think there's a reason for this and that Mooi and Gill are being
unfair to molecular systematists. I'll try to explain what I mean, but
first understand that I'm not talking about homology at the molecular
level. When you compare sequences, homologies are alignment columns. The
column is the character, and the nucleotide in that column is the
character state. These homologies cannot be inferred in the same way
that morphological homologies can be. Instead, they are either created
in toto by MSA optimisation, or (in the case of dynamic optimisation)
inferred and tweaked by an algorithm that is simultaneously looking for
a tree meeting certain criteria.

OK, so the characters (columns) are now defined. Now what? Well, in
morphology structures don't evolve from nothing, and changes (even when
drastic) have observable roots in pre-existing structures. In sequences,
character state changes aren't like that. It's possible for any
nucleotide to change into any other nucleotide. There are constraints,
but we don't know what they are at any given position. Instead we have a
zoo of models which give transition probabilities (that's mathematical
transition, not transition/transversion transition), and fudge factors
for autocorrelation along the sequence.

Do you agree?
-- 
Dr Robert Mesibov
Honorary Research Associate
Queen Victoria Museum and Art Gallery, and
School of Zoology, University of Tasmania
Home contact: PO Box 101, Penguin, Tasmania, Australia 7316
03 64371195; 61 3 64371195
Webpage: http://www.qvmag.tas.gov.au/mesibov.html

_______________________________________________

Taxacom Mailing List
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom

The Taxacom archive going back to 1992 may be searched with either of
these methods:

(1) http://taxacom.markmail.org

Or (2) a Google search specified as:
site:mailman.nhm.ku.edu/pipermail/taxacom  your search terms here