I would be interested in the answer to the following question by Keith.
We currently use the blastcl3 client in Perl scripts to do this sort
Also, a question re Overlap_right and Overlap_left:
There is asymmetry here. The ace_4.7g models have an Int for Overlap-right
but not one for Overlap_left. This Int, the overlap to the next 'leftward'
sequence, eg a cosmid, is apparently used in Richard's LINK script to generate
But ... is there more to this?
Why would users also not be interested in the overlap to a 'rightward'
Are Overlap_right and Overlap_left 'magic tags' in some sense?
>On Tue, 9 Nov 1999, Richard Durbin wrote:
>>> Hello Keith,
>>>> Aha, you ended up with AGR. Congratulations.
>>Thankyou, I started last Monday (which was the same day that I finally
>submitted my thesis).
>>Thanks for the help. I won't pretend that I've got to grips with Acedb
>yet (certainly not within a week). I will hopefully be attending the
>ACEDB course in January and that will help.
>>There is no urgent need for me to sort this out for the Arabidopsis
>database as there is still a lot of sequence information to come before it
>will be necessary to link things together.
>>One more question though. The overlap information stored in the
>Overlap_left and Overlap_right tags...does this come from the comment
>field of the GenBank/EMBL entries? Without a lof of BLAST searchs I don't
>see how else you piece together which sequences overlap. I'm therefore
>uncertain of how applicable this is to A. thaliana as from what I have
>seen so far, very few EMBL entries have information about how they
>overlap...which begs the questions as to how our database knows that they
>*do* overlap in the first place.
>>I shall try and refrain from asking you too many questions before I've
>learnt a little more about ACEDB.
>>Thanks again for your help.
>>~ Keith Bradnam - Developer, Arabidopsis Genome Resource (AGR)
>~ University Park, University of Nottingham, NG7 2RD, UK
>~ Tel: (0115) 951 3091 Web: http://synteny.nott.ac.uk/>>---