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

Re: Shaking up the bird family tree



> Interestingly, Metaves and Coronaves are found, but not
> supported, perhaps 
> because of the pesky hoatzin. The interesting part is that
> only two of the 
> 19 loci find it, but, in spite of this, the other 17 do not
> manage to 
> overcome this signal.

Found, but the branches leading to them and to each of its principal groups are 
even shorter than in the rest of the "backbone". So the signal is not 
overruled, but worn down to almost nothing. It would be good to know where the 
signal comes from. If you include beta-fibrinogen, you will find this apparent 
dichotomy; that much is known. If you for example compare mt-genomes, the 
dichotomy disappear without a trace. And to quote the original paper:

"Metaves and Coronaves are broadly distinguished by four indels: a five-base 
deletion in Coronaves, a 15-base deletion in Coronaves plus Syrrhaptes (one of 
two genera of sandgrouse, both studied), a 21-base deletion in Coronaves plus 
ostrich or insertion in Metaves plus Galloanserae, and a four-base deletion in 
Metaves plus bustards."

So "a 21-base deletion [...] or insertion" is something that *did* happen 
convergently, and the last deletion/(if the dichotomy is bogus) insertion shows 
that frame-shift indels *did* happen convergently too. 

Another way to put the original rationale for the Metaves would be "presently 
supported by at most one unambiguous apomorphy in a character type where 
extreme convergent evolution is known". And it is probably not a good idea to 
make much of such "clades" either (Would anyone here confidently write a 
16-pages paper proposing clades based on a *morph* analysis yielding such a 
situation?).

The beta-fibrinogen signal is known to be VERY strong; it can override 5 times 
the amount of data contradicting it. But as you can see, it is not very robust. 
It is strong simply because it's whole blocks of identical bases missing or 
inserted in a single instant, overruling single-base changes across the entire 
sequence. OTOH, there is sufficient evidence that certain loci "attract" 
transposons: the "21bp deletion" may just as well be a widespread 21bp 
insertion (you cannot tell in the scope of a cladistic analysis).

So a single evolutionary indel event changes, for the purpose of this analysis, 
several characters. Whereas a single point mutation event changes a single 
character.

Changing the alignment algorithm (in BLAST terms: using PAM and not BLOSUM) 
would probably prevent you from ever finding this supposed dichotomy. 
Alternatively, one might treat indels as a single character each - which they 
technically *are*. In that case too, the signal would be easily overwhelmed.

The problem with molecular phylogenetics in deep time, at its present state, is 
that the belief (founded upon its power in not-so-deep-time analyses) in its 
capabilities is vast, but the knowledge about what precisely the limitations 
are is marginal. doi:10.1080/10635150590947843 gives a good indication of the 
direction research has to take before there can be any chance to resolve such 
uncertain radiations. 

And if I am correct, the neoavian "backbone" is characterized by drift and 
random lineage sorting as heavy as can be expected in lineages that barely 
manage to make it through a catastrohic mass extinction... all evidence 
suggests that it is simply wrong to assume "normal" molecular evolution in that 
region of the neornithine tree.

In conclusion: we need bones (or analyses thereof), because the bases simply 
won't cut.


Eike


      __________________________________________________________
Gesendet von Yahoo! Mail.
Dem pfiffigeren Posteingang.
http://de.overview.mail.yahoo.com