chloroplast and other genes (was Lucy in Newsweek)
deepreef at BISHOPMUSEUM.ORG
Thu Apr 8 08:54:42 CDT 2004
> it remains possible - and to me probable - that much
> misinformation also exists there, in the form of homoplasious
> matches (parallel evolution).
Yes, of course -- that is a given. That is why I have also maintained that
simply knowing the whole genome won't help us much. The REAL missing
ingredient is the *understanding* of the genome -- which means the ability
to generate algorithms that recognize and take into account the homoplasy.
If it proves impossible to develop algorithms that can do this reliably,
then the "asymptote" will be that much lower (in other words, the
information will simply not be retrievable). To whatever extent the
information exists in the genome (and this includes the information needed
to avoid being misled by misinformation), I am convinced that we will
eventually be able to extract that information (and filter the
misinformation). The limitation will be in how much information about
phylogenies has been lost -- obliterated -- forever.
> It has been my experience that, in morphological data
> (sens.lat), a minimum of some 15-20% of matches are homoplasious.
> A few molecular studies known to me (that provided the necessary
> data - which is not usual!)did little if at all better! Is there
> any reason to believe that using enormous numbers of codon
> matches is going to improve on those figures?
Again, I emphasize that I am *NOT* simply talking about applying the same
algorithms to zillions more base pairs. I'm talking about reaching a
threshold of understanding about the genome that allows us to extract all
the information it contains. Someone with a rudimentary understanding of
the English language who tries to read Shakespeare will probably interpret
all of it literally, and won't likely pick up on all the metaphors. That's
how I see our current ability to read the genome -- we have access to only a
few pages, and we are trying to interpret each word literally. When we have
access to the story in its entirety (full genomes of all organisms), *and*
we understand enough about the language that we can recognize the metaphors
(recognizing homoplasy being only one tiny example), then we will have a
much richer understanding of the true meaning of the story.
> I was once persuaded to write a paper on "Information" in
> taxonomy. It convinced me that the concept is multifacetted,
> slippery, easily misused, and generally downright dangerous.
I WHOLEHEARTEDLY agree!
> Let's just stick to "data" (and, please, as a plural noun!)
But science is about more than just gathering data -- it is also about
interpreting data. And that's what seems to be the crux of the points that
both of us are trying to make.
Richard L. Pyle, PhD
Ichthyology, Bishop Museum
1525 Bernice St., Honolulu, HI 96817
Ph: (808)848-4115, Fax: (808)847-8252
email: deepreef at bishopmuseum.org
More information about the Taxacom