[ensembl-dev] Multiple 'paths' in assembly table?

[Dan Bolser]

Thanks Sharon,

I got a similar reply off list.


Thanks all,
Dan.

On 4 September 2012 18:13, Sharon Wei <weix at cshl.edu> wrote:
> I ran into the same problem before and from my experience, the only solution
> is to use dummy scaffolds, otherwise the ensembl API won't work correctly.
>
> Sharon
>
>
> On 8/31/12 12:20 PM, Dan Bolser wrote:
>>
>> Hi,
>>
>> Some of my contigs map to chromosomes via scaffolds, and some are
>> included in chromosomes directly. This means that there are multiple
>> (two) paths from contig to chromosome. No matter what I put in meta
>> (assembly.mapping), it seems I can't project both types of 'seqlevel'
>> to 'toplevel' correctly.
>>
>> Either I can project only those contigs found in chromosomes via
>> scaffolds OR only those contigs found in chromosomes directly, but not
>> both types with one set of assembly.mapping meta keys.
>>
>> Is it possible to correctly project either type of contig with one set
>> of assembly.mapping meta keys?
>>
>> I need this to correctly run the 'set_toplevel.pl' script.
>>
>> I can get round the problem by adding dummy scaffolds for all my
>> contigs that are not part of real scaffolds, but this introduces
>> thousands of dummy seq_regions and assembly entries... Obviously, if
>> this is the only way to do it, that's what I'll do.
>>
>> Finally, how much time does the assembly.mapping meta key save?
>> Thinking about it, I can't convince myself that you save any extra
>> queries by having this information (but I haven't looked at places in
>> the API where it may be used).
>>
>>
>> Thanks for suggestions,
>> Dan.
>>
>> _______________________________________________
>> Dev mailing list    Dev at ensembl.org
>> List admin (including subscribe/unsubscribe):
>> http://lists.ensembl.org/mailman/listinfo/dev
>> Ensembl Blog: http://www.ensembl.info/
>
>
>
> _______________________________________________
> Dev mailing list    Dev at ensembl.org
> List admin (including subscribe/unsubscribe):
> http://lists.ensembl.org/mailman/listinfo/dev
> Ensembl Blog: http://www.ensembl.info/