Re: [SO-devel] GFF3 Specification

[Joachim B. <joa...@gm...>]

On 2013-02-26, at 4:37 PM, Peter Cock <p.j...@go...> wrote:
> The fact that Joachim Baran commented that he'd noticed
> this in GVF data from Ensembl suggests this is more than
> just a hypothetical usage, and it could be premature to ban
> in the new revision of GFF3.
  It should be closely reviewed though. Ensembl's GVF files do not adhere to the GVF specification very well. For example, their variant sequences contain attributes such as "Variant_seq=INSERTION" and "Variant_seq=(11 BP INSERTION)".

  Either there is a lack of expressiveness in GVF that forces Ensembl to violate the specification in that way, or there is a lack of clearness in the specification's text. (Well, or I get the specification wrong myself...)

Best,
Joachim