DNA and identifications

[Regis Courtecuisse]

Dear all,

    The necessity of having a complete set of DNA sequences for compariso=
n is
of course one important problem (far from being solved as Thomas suggests=
).
    But the accuracy of the identifications made to create this DNA seque=
nces
database is another crucial point. For such a purpose, taxonomists (in th=
e
traditionnal meaning) are definitely necessary to make the job.  (Who els=
e ?)
    I know actual cases where DNA-based identification came to a wrong
conclusion because the reference material was erroneously determined by a
non-specialist. Of course, molecularists are unable to judge the accuracy=
 of a
taxonomical identification and are quite confident in their molecular met=
hod.
What they find while comparing the DNA of an unknown specimen with the
reference data set MUST be the correct name. The problem is that we still=
 need
a preliminary correct identification to make the method simply credible.

    The present balance between the number of reliable taxonomists and th=
e
legions of molecularists make me very pessimistic about the very future o=
f
molecular biology. When taxonomists will become extinct (if the tendency =
still
runs to this misfortunate issue), molecularists will really feel alone...

    R=E9gis

"Thomas G. Lammers" wrote:

> At 11:28 AM 01-15-99 CST, you wrote:
>
> >To the gene jocks out there,
> >Question that came up recently and is beyond my
> >expertise... if you had a piece of plant could you
> >identify it to species just by DNA-type analyses?
> >The only information you would have is that the
> >piece was from a plant but you had nothing further, no
> >locality, no data on genus or family or even if gymnosperm
> >of angiosperm. (I am fully aware that once you know what
> >species you can identify the indivudual from whence the
> >piece came.) I was told there is a lab in Arizona that can
> >do this type of ground zero identification but haven't found
> >anything on the web. Anyone know about this lab?
>
> I'm no molecular johnny,  but identification such as this would seem to=
 be
> predicated on having known sequences to compare against.  I would suspe=
ct a
> very tiny fraction of the world's species have been sequenced for any o=
ne
> gene and the results communicated to GENBANK.   Better results could be
> obtained much quicker and much more cheaply with a good herbarium speci=
men.
> : - P
>
> Thomas G. Lammers
>
> Classification, Nomenclature, Phylogeny and Biogeography
> of the Campanulaceae, s. lat.
>
> Department of Botany
> Field Museum of Natural History
> Roosevelt Road at Lake Shore Drive
> Chicago, Illinois 60605-2496 USA
>
> e-mail:    tlammers at fmnh.org
> office:          312-922-9410 ext. 317 (voice-mail)
> fax:                312-427-2530
>
> -----------------------------------------------------
> "The most important thing to learn is how
>    to teach yourself.  Everything else ...
>    is just detail."
>                                   -- Patrick Gibson

R=E9gis Courtecuisse
D=E9partement de Botanique
Facult=E9 des Sciences Pharmaceutiques et Biologiques
BP 83 ; F-59006 Lille Cedex