GenBank & Taxonomical Nomenclature/identification

[Stuart G. Poss]

Another example where if we had coordinate data structures for molecular
and legacy databases, an electronic agent method could flag the
appropriate parties that a change has been made (ie an update of an
identification in a archival database could flag the GenBank database
"curators" that a new identication has been made at the archival location
or flag an "expert" that a sequence on their favorite taxon has been
posted).  If both databases recognized the digital signature of experts,
then the entire system could be able, via yet another agent method,  to
instantaneously assess the quality of the identification data and, in
principle, even track the statistical accuracy of the experts themselves.

We really need to reevaluate how we can get newer network-centric
programming (primarily in Java or Perl) to do the chores that humans are
not too good at, but at which computers excel.  We often really don't
need to pass the responsibility for maintenance on to another human when
a computer will be "happy" to do it for us, if we just instruct it
properly.  Cross discipline dialog like that being engaged in now, could
do much to stimulate us to think about precisely how this might best be
done.

Is the data structure for the GenBank database freely available?

Stuart Poss


Carol Hotton wrote:

>
> If (for example) a species is misidentified in a sequence record after
> it is deposited in GenBank, the record has to be updated manually by
> the GenBank indexing staff.  We generally tell the indexer changing
> the record to insert a note in the sequence record: /note="submitted
> as <oldname>.  But unfortunately this has not been done consistently,
> and many sequence records, especially older submissions, often lack
> this information.
>

>
> Hope this clarifies things a bit...
>
> Carol Hotton
> GenBank taxonomy staff
> >Carol,
> >Can you also address the issue of retaining the history of
> >identification changes which are made to a sequence (aka a
> >"specimen")?  It's important to know what name people thought
> >they were dealing with when they analyzed data and drew
> >conclusions about the species (i.e., about the name that was
> >associated with the sequence/specimen at the time they retrieved
> >the data for analysis).
> >Peter
> >
> >On Thu, 27 Jul 2000, Carol Hotton wrote:
> >...cut, cut...
> >> Submitters are indeed allowed to update any data relating to their
> >> sequences submssions at any time, including of course, taxonomic
> >> revisions, changes in classification, voucher information, or
> anything
> >> else related to the deposited sequence accessions.
> >
> >
> >> .... The five of us are responsible for maintaining
> >> an up-to-date classification of the entire range of living
> organisms,
> >> and we need all the help we can get!
> >
> >Maybe we're not talking about the same objects? I thought the
> >reference was to the specific epithets assigned to a _specimen_,
> >and whether or not those "determinations" were able to be
> >revised, and also whether or not the history of the revisions is
> >kept.
> >
> >Peter
> >