[Bioperl-l] bond() operator

[Hilmar Lapp]

I guess we should come to a decision here relatively soon. It is not 
nice if people want to parse through a file and one entry that throws 
an exception just stymies them. Ideally we would be able to include a 
fix, if any, in 1.2.1 (Ewan?).

These are the options as I see them. No claim of completeness though.

a) Do nothing and wait it out until NCBI uses a saner way to annotate 
this.

b) Similar to a), but change the exception to a warning. (whole feature 
would be ignored)

c) Similar to a), but ignore the enclosing bond() for the time being, 
without even warning. (feature itself would not be ignored)

d) Make every effort to retain the bond() operator as the feature's 
tag/value. (this may require some nasty lines of code, but is doable)

e) Add genbank_loc_operator() to SplitLocation and store the value of 
the operator there.

If you see more options please post.

Other than that, please vote.

	-hilmar

On Tuesday, March 4, 2003, at 01:54  AM, Heikki Lehvaslaiho wrote:

>> From Guenter Stoesser, , EMBL Database Coordinator:
>
> - 'bond' is not a collab feature, and has never been discussed with or
>    mentioned to the collaborators (EMBL & DDBJ)
> -  this seems to appear in a Refseq entry - Refseq activities are not
>    part of the collaborative effort and basically NCBI can do there
>    whatever they want - besides the confusion it creates with the user
>    community, since navigating via Entrez does not make it clear to the
>    user that Refseq records are not part of the DDBJ/EMBL/GenBank work
>
> --Guenter
>
>
> Mey it is time make clearer distinction between GenBank and RefSeq
> parsers?
>
> 	-Heikki
>
>
> On Mon, 2003-03-03 at 21:02, Hilmar Lapp wrote:
>> On Monday, March 3, 2003, at 08:20  AM, Ewan Birney wrote:
>>
>>> Well... is this a bug?
>>>
>>>
>>> http://www.ebi.ac.uk/embl/Documentation/FT_definitions/
>>> feature_table.html#operators
>>>
>>> bond is not in there.
>>>
>>
>> Right. That's why I sent a grumbling email when this first came up
>> whether Genbank wants to repeat the same mistake (mistake IMHO 
>> anyways)
>> it made a while ago with annotating polymorphisms, namely to fold
>> annotation into the location string.
>>
>> With polymorphisms, they used the dreaded replace('A', 100) syntax.
>> Meanwhile they corrected this and it's now a feature with key
>> 'variation' and standard location syntax.
>>
>> Having annotation and location folded into the same string really
>> creates a whole mess in bioperl with regard to parsing this out in a
>> structured manner, and even worse, round-tripping. So, quite honestly,
>> I do hope that NCBI backs off of that, as they did for replace(). But
>> maybe I'm missing something.
>>
>> 	-hilmar
> -- 
> ______ _/      _/_____________________________________________________
>       _/      _/                      http://www.ebi.ac.uk/mutations/
>      _/  _/  _/  Heikki Lehvaslaiho          heikki at ebi.ac.uk
>     _/_/_/_/_/  EMBL Outstation, European Bioinformatics Institute
>    _/  _/  _/  Wellcome Trust Genome Campus, Hinxton
>   _/  _/  _/  Cambs. CB10 1SD, United Kingdom
>      _/      Phone: +44 (0)1223 494 644   FAX: +44 (0)1223 494 468
> ___ _/_/_/_/_/________________________________________________________
>
>
-- 
-------------------------------------------------------------
Hilmar Lapp                            email: lapp at gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
-------------------------------------------------------------