Geoff Read wrote:
>> I know of instances where new molecular studies has found the
>> misidentification in the old work, but are there cases where an actual
>> taxonomist has re-examined the voucher and found the misidentification simply
>> by eyeballing it?
>>>> None has been mentioned to me. So, strike one against the usefulness of vouchers?
>I think it is too early for that to happen. After multiple sequences of
(supposedly) same species accumulate on GenBank, examining vouchers will
be important to decide whether a mis-indentification was the explanation
of inferred genetic variation
>> Here is an interesting complex case in freshwater limpets involving a
>> misidentified GenBank sequence:
>>>> Walther, A. C.; Lee, T.; Burch, J. B. ; Foighil, D. O. 2006: Acroloxus
>> lacustris is not an ancylid: A case of misidentification involving the
>> cryptic invader Ferrissia fragilis (Mollusca: Pulmonata: Hygrophila).
>> Molecular Phylogenetics and Evolution 39(1): 271-275.
>>http://dx.doi.org/10.1016/j.ympev.2005.08.014>>>> Even more interesting is that the misidentified entry (AY577462) as A
>> lacustris is still seemingly uncorrected in GenBank. Why is this? Have I got
>> this right? It seems very strange.
>> See http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?46451261:NCBI:6982153>>>> There does not seem to be a process for corrections or warning annotations to GenBank identifications by subsequent workers,
> and the very notion of original misidentification may be an alien concept to the working processes there. So one shows the voucher is misidentified but there no obvious path to get this recorded? Is this correct? Strike two against the usefulness of vouchers?
>I am not sure, but an inquiry can be sent to GenBank. Again, GenBank's
policies are not set in stone, they are evolving, and if we can show
that such a path is needed, I am sure it will be created
> "Revisions or updates to GenBank entries can be made by the submitters" Right, only by the submitters, but they don't do it. They just hold tight, take the heat, and pretend (in the GenBank record) they didn't stuff up. Strike three and out for usefulness of vouchers in practice?
>Again, as I said above - with multiple diverse sequences attributable to
one species such a stuff up will be really difficult to hide
> Ps - a case of switched identities between the sequences of 2 vastly different worm taxa has been mentioned to me privately. One just happens to be a taxon I'm interested in. Nice to know - but it should be GenBank which told me.
>What does BLAST tell in this case? And why to not to correct this
mistake now, once it is known?
Wormly,
Lena