[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

RE: New paper on Neoaves

Christopher Taylor wrote:

    To correct something Tim said quite a bit earlier, Ericson et al.
did find Palaeognathae to be 'monophyletic', but this wasn't shown as
such in the paper because Palaeognathae is the outgroup.

Ericson et al. state that two palaeognaths (rhea, kiwi), a galliform (megapode) and an anseriform (screamer) were "used as outgroups". The Materials and Methods does not say if these four taxa were defined as such (a priori) when the authors were setting up the analysis. I've never done a Bayesian analysis, but PAUP gives you the option of defining outgroups(s) beforehand.

There can be such biases in morphological data, too - witness the determination of loons and grebes to stick to each other due to characters related to foot-propelled diving.

Definitely. But as you can you see from your example, it is relatively easy to identify the source of potential homoplasy in morphology. In this case, it is characters related to foot-propelled diving shared by grebes and loons. Similarly, penguins and plotopterids (to use David's example) might be being pulled together by characters associated with their underwater lifestyle (as interpreted by Olson), or they could reflect a common ancestry (as Mayr proposed more recently, based on his phylogenetic analysis). This is a tricky business, sorting convergence from common ancestry - but I like to think that, given enough characters, the latter wins out. But the potential sources of homoplasy in molecular phylogenies are more subtle, and require further analyses to identify.

What I am not aware of any studies on so
far is whether any such biases that may be taxon-variable (such as codon
bias) are found throughout the genome of any given species, or whether
different parts of the genome may differ in these biases.

For higher eukaryotes, I believe the latter is true (at least for the isochores of the nuclear genome). Compositional bias may affect the entire genomes of prokaryotes, as well as mitochondria and chloroplasts.

Stringing sequences together from different genes may obscure differences in local biases,

It should remove DNA-level localized base-compositional biases, but some other biases may persist (especially with certain amino acids).

Also, a mitochondrial genome should be effectively regarded as a single unit, which may not
necessarily be the case with the nuclear genome, as different sections may have different evolutionary histories.

I think this is largely true, although I would specify that the light and heavy strands of the mitochondrial genome should probably not be treated the same way.

(from another posting):

These characters may be ancestral for deuterostomes - as demonstrated by the fossil record and morphology of modern taxa.

Nothing would surprise me. Nevertheless, I think there is an unfortunate tendency by some studies to re-write morphological evolution through the prism of molecular-based trees. When a morphology-based tree and a sequence-based tree conflict, the instictive reaction is to assume that the morphology-based tree is wrong. For me, the conflict may lie with faults on both sides.

I've got nothing against proposing new and exciting evolutionary scenarios based on novel topologies given by sequence-based analyses. However, I see some 'shoehorning' going on, particular with these more heterodox scenarios.

Cheers

Tim