[Taxacom] Dark taxa: GenBank in a post-taxonomic world

[Curtis Clark]

On 2011-04-13 07:54, Richard Zander wrote:
> This is my usual devil's advocate stance, but given some attention to limitations, okay certainly some research is possible on just a raw big set of sequences. It should be grounded in expressed traits and environment, somehow, eventually, as Bob Mesibov says.
I'm advancing a different candidate for sainthood than the one you are 
arguing against, and I'm not necessarily disagreeing with either you or 
Bob. What I'm suggesting is that Rod's analysis tells us nothing novel 
or surprising; that the observed pattern is not the result of a change 
in attitudes toward formal names, but rather the playing out of forces 
already at work. The only "attitude" parameter that's necessary for my 
hypothesis is that, in general, scientists would rather be recognized 
than not.

Your point about whether the set is finite or infinite is an interesting 
one. For it to be infinite, the rate of occurrence of novel changes (as 
contrasted to back-mutations to characterized states) must exceed the 
rate of sequencing. The latter is increasing almost exponentially, so I 
would expect that the curves will eventually intersect, and the set is 
finite. Whether that makes any difference in the near term is another 
matter.

-- 
--
Curtis Clark
Cal Poly Pomona