Fwd: Re: [TAXACOM] genetic vs morphological trace of phylogeny

[A.P. Jason de Koning]

Since I've made my feelings on John's approach to cladistics clear in the past, I don't want to plunge too deeply back into this thread...nevertheless, a few points regarding genomes and human-ape relationships:

A major benefit of having complete genome sequences is the ability to exhaustively score structural characteristics such as retroelement presence.  As Ken and others have pointed out, retroelements probably make for particularly reliable sources of phylogenetic information; some have even labelled these as being nearly "homoplasy free" characters...you've got to admire that kind of hubris :)  (I personally think we need to know a little more about the molecular biology of insertion, since there does seem to be some statistical support for biased insertion of certain types of retroelements.)

Apropos to John's crusade for a Schwartz-orang sister relationship, a recent study by Mark Batzer's lab scored the presence/absence of 153 known Alu Ye subfamily members across multiple individuals from 8 primate species (human, 5 apes, 2 monkeys), by PCR-amplification using primers designed against the human genome and against primate DNAs in GenBank.  Their analysis unambiguously resolved the conventional ((H,C),G),O) relationship.  For John's edification, I also reanalysed their dataset as follows:

Performing an unconstrained, exhaustive search on their dataset, the conventional relationship is resolved by a tree with 124 steps using Dollo parsimony.  The consistency index was something like 0.999.  With John in mind, I forced a human-orang sister relationship and performed an exhaustive search with that constraint.  The shortest tree supporting this hypothesis was 166 steps, and was strongly excluded by all relevant statistical comparisons against the optimal unconstrained tree, with P < 0.0001.  (I can post details or data if anyone wants.)

While their dataset cannot be exhaustive, and could possibly be complicated by ascertainment bias because of using mostly human-designed primers, there doesn't appear to be any reason to infer directional bias.

You can find a link to the source paper at:
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14561894

Cheers,
- Jason

__________________________________________________________________________
  A.P. Jason de Koning, Doctoral student        Email: apjdk at albany.edu
    Department of Biological Sciences             Lab:  (518) 442-4347
    University at Albany, SUNY                    FAX:  (518) 442-4767
    1400 Washington Ave., Albany NY 12222, USA  Mobil:  (518) 210-4504
__________________________________________________________________________