DNA goggles

Leo A. Martin leo@possi.org
Thu, 31 Jan 2013 21:22:16 PST
>> DNA sequencer goggles are not (currently?) good enough to detect variants
>> and subspecies, so many people whose DNA goggles are stuck on their heads [hope/ suspect
>> /
>> assume / believe / think / behave as though / wonder whether] variants and subspecies
>> do not actually exist.

Nhu replied

> Not true, Leo.

So, if I threw you a box of 500 plants that morphologists said belonged to a number of
different species and genera, and you chopped them up without looking at them to see
their morphology, and isolated their DNA, you would be able to segregate them into the
correct species and genera as determined by the morphologists?

No more information, by the way, such as How many genera in the box or How many species
in the box or How many taxa in the box or Are there any duplicates?

Of course I realize entire genomes can be sequenced. We think (hope?) it's without error
and without too much trouble. But the problem is a human has to make some kind of
decision about what is a species and a human has to make a decision of how this array of
DNA sequences and gene products maps to species already described. The data you describe
don't sort themselves, don't interpret themselves, and there's no
autospectrospeciesometer. If you chop up a bunch of plants from one location for
sequencing and somebody adds a few non-related outsiders without you knowing - or some
algae spores float in before the PCR reaction - you will think the population has more
DNA sequences and gene products than it really does. And you can't tell whether you have
extra plants thrown in without somebody looking at them, so the DNA sequences become a
proxy for morphologic criteria.

Convergent evolution may conceivably result in similar DNA sequences or gene products
from varying origins.

And as for identifying organisms in mud from their DNA - How would anybody know there's
just one organism there?

Others have pointed out this field is young and people rarely in fact sequence entire
plants, because nobody's found out how to monetize it. But even with perfect sequencing
of all the plants anybody's found, humans will decide what will be called a species. And
throwing out all previous families, genera and species in favor of a new classification
system based on presumed DNA evolutionary lineages seems like the opposite of using
Ockham's razor. We tend to minimize ancient science because it seems so little advanced
compared to ours - but the people doing the science in the 1700s were just as brilliant
as the best scientists now and no more prone to mistakes.

I'm not by any means a Luddite saying DNA research is valueless. I'm saying the field is
so young we don't have any idea what we know nor how it maps to what we already know,
and I'm saying conclusions may be rushed to publication without enough supporting data,
and I'm saying the first thing I was taught in experimental design class was how easy it
is for scientists to unknowingly deceive themselves.

Leo Martin
Phoenix Arizona USA

More information about the pbs mailing list