[Taxacom] Clade age (was: Taxacom Digest)‏

Jason Mate jfmate at hotmail.com
Thu Nov 10 18:14:31 CST 2011






I apologize for dropping out for a while but I had to undergo minor surgery after my trip and as everybody has moved on there is little point in pursuing the previous topic. Live and let live! In regards to Ann´s questions it boils down to mutation rates. There are several oft quoted numbers (averages) in the literature. Something in the range of 0.7-4/site per 10(power of 9) years are the general ballpark figures (although for Drosophila I have seen as high as 14?). These are only general average figures. You have to consider the inherent biases (compositional biases, positional biases, structural, etc), so your mutation rates are only a rough approximation for a particular bp.
If your gene of choice fails to recover any differences in a group of species I suggest trying another commonly used gene. Trial and error is the only way to find out which gene shows promising variation (and it could still be the wrong sort!). I recall a presentation on Curculio phylogeny which encountered a similar problem with certain species pairs, even when the rest of the group showed "normal" variation. The article is to be found in this link http://www.ncbi.nlm.nih.gov/pubmed/15223041, and I imagine that they discuss it.
In regards to the second topic, it is a problem that affects systematics in general and not just cladistics (although I accept Richard´s comment that some hard core cladists would have problems accepting this scenario). Speciation is hardly clean but if you assume, for the sake of argument, that a range was "instantly" divided into separte ranges that then evolved into separate species then you would most likely end with a hard polytomy or at best a poorly supported cladogram that may not even represent the biogeography of the group (still a polytomy). But, as there is no pattern of biogeographical subdivision to uncover anyway, who cares?! However, if the range was broken "gradually" (from the perspective of mutation rates) then it is likely that some of the splits will be recoverable. Of course the resulting phylogeny may still be rubbish. My only questions in this regard are if the gene that Lynn mentions shows no variation at all or is the variation just "noise" (I don´t like to call it noise as if is information after all, just the wrong kind for your question).

Best
Jason
 		 	   		  


More information about the Taxacom mailing list