GenBank Errors

Michael J. Cinkosky michael at DOMAIN.LANL.GOV
Fri Oct 18 13:32:15 EST 1991

    I do not know what sort of errors you detected in GenBank
or who you spoke to that indicated that we did not want to
implement software for detecting and/or correcting such errors,
but I assure you that we are very interested in, as well as
actively involved in, producing such software.  Since our 
conversion to a relational system (one reason for our dramatic
improvement in performance) we have had far more opportunity
to design and implement these essential software checks.  The
number of problems in the current database, while dropping
quickly, is of great concern to us.  Our goal is nothing short
of producing an error-free database with at least our current
turn-around time.  Currently we have targeted what we perceive
to be the most glaring errors and we are devoting both programming
and review effort to correcting them.
    Perhaps the message that you interpreted as meaning "we are
not interested in implementing automatic checks" was really, "we
are not currently implementing this particular automatic check
because we have chosen a different priority."  However, if you
(or anyone else) has suggestions for checks that you would like
to see done on the data (or even just input on which errors you
think are most important and should be addressed first), we want
to hear from you.  If nothing else, could you please at least
tell us what errors you are talking about so that we all can
benefit from your careful examination of the database?
    We regularly process more data each week than we used to
handle in an entire year, and we do it with far fewer errors
than ever before.  But we, like you, will not be satisfied until
we have eliminated all the errors that we can from our work.
I am sorry that you have gotten the impression that we are on
opposite sides of this issue; we are not.  I hope that you, like
many others before you, will help us in our efforts at quality
control by continuing to point out errors as you find them (I
think we have a good record of promptly correcting such errors)
and telling us about any ideas that you have for automating the
detection/correction process.  

Michael Cinkosky
Computation Domain Leader
Los Alamos National Laboratory

More information about the Bioforum mailing list