[Taxacom] shared derived molecular similarities

[John Grehan]

-----Original Message-----
From: Sergio Vargas [mailto:sevragorgia at gmail.com] 
Sent: Monday, April 04, 2011 2:51 PM
To: John Grehan; taxacom at mailman.nhm.ku.edu
Subject: Re: shared derived molecular similarities

>Even with coding genes the genes are not necessarily the same length so
>one has to invent the homologies even before theorizing the
replacement.

well, I think, in principle, the problem is not so different from
establishing homology
between morphological parts when one or more sections have been deleted.


Agreed - when one is comparing different numbers of entities it can be
as problematic to figure out (such as with teeth) what is homologous or
not by position alone. But morphological comparisons are not limited to
this problem, and even the positional question may be addressed by
looking at individual morphology of the entities to see if there is any
comparative evidence. Can't do that with sequence alignment which
creates homologies that cannot be seen in nature (they are metaphysical
as far as I am concerned).

Also, the alignment issue has been overemphasized; is only a real
problem with certain non-coding
sequences or part of some "genes" (like the loops of ribosomal RNA).

Seems to me that alignment is employed quite often in coding genes.

>And in practice the outgroup is often so limited in taxonomic
representation as to almost be meaningless.

well this one is easy... add more outgroups... or?


Yes add more outgroup with substantial amount of diversity and species,
or add more representatives in the outgroup. In theory the outgroup
should be completely represented and that provides the strongest
corroboration of a derived state for the ingroup. For example, it is
often the practice for human-great ape comparisons to use an outgroup
comprising the gibbons with only 11 species and fairly homogenous to
boot. This occurs in morphological as well as molecular comparisons. But
even then it is usually one or two gibbon species. The next outgroup is
nearly 200 species of OW Monkeys, yet when molecular studies use monkeys
the number of representatives is usually 10 or less, and often for
particular genera. It would be interesting to see how many molecular
homologies supposed for humans and chimps would stand up with all OW
monkeys and gibbon species included. 

John Grehan