[SO-devel] Re: new/changed SO terms for FlyBase

[Karen E. <ei...@fr...>]

Changes requested by Gillian:

See comments

On Feb 8, 2006, at 3:52 AM, Gillian Millburn wrote:
>
>
> Here they are (have discussed these changes with Michael)
>
> 1.
>
> [Term]
> id: SO:0000042
> name: pseudogene_attribute
> def: "An attribute of a pseudogene (SO\:0000336)." [SO:ma]
> is_a: SO:0000009 ! gene_class
>
>
> please could the name of this term be changed to :
>
> name: pseudogene_class
>
> It reads nicer for biologists, and makes more sense if when annotating 
> a
> genes file you have a gene you know is a pseudogene but not which
> subclass its is.
>
> I guess that the definition will need changing/deleting too, as it
> doesn't make much sense to me.
>
I do not think that SO is doing the best job with pseudogenes. Will get 
back to you about this when I see how the other cross product terms are 
received.

>
> 2. suggestions for tidying up the "engineered_gene" section.
>
> the section of terms that has "engineered" in them needs a bit of
> rejigging to be able to cope with all the transgenic situations that
> happen in the fly.
I have made an attribute "engineered". Now we can make cross product 
terms of engineered features as needed.

>
> a. not all engineered bits in flies are a gene, but just a sequence, so
> we need a couple of new generic terms to accommodate that:
>
> [Term]
> id: SO:NEW
> name: engineered_sequence
> is_a: SO:0000400 ! sequence_attribute
>
[Term]
id: SO:0000804
name: engineered_region
namespace: sequence
synonym: "engineered sequence" RELATED []
intersection_of: SO:0000001 ! region
intersection_of: has_quality SO:0000783 ! engineered

>
> [Term]
> id: SO:NEW
> name: engineered_foreign_sequence
> is_a: SO:NEW ! engineered_sequence
>
[Term]
id: SO:0000805
name: engineered_foreign_region
namespace: sequence
intersection_of: SO:0000001 ! region
intersection_of: has_quality SO:0000783 ! engineered
intersection_of: has_quality SO:0000784 ! foreign

>
> b. it also makes sense to delete the following SO term:
>
> [Term]
> id: SO:0000283
> name: engineered_foreign_transposable_element_gene
> is_a: SO:0000111 ! transposable_element_gene
> is_a: SO:0000280 ! engineered_gene
>
> since this SO term seems to be combining two separate concepts -
> foreign-ness and TE_gene-ness and in general SO tries to get away from
> that.
>
> it can be accommodated by the following two terms:
>
> engineered_foreign_gene ; SO:0000281
> transposable_element_gene ; SO:0000111
I am not sure if this makes a lot of sense, but I made this term a 
cross product too.

[Term]
id: SO:0000283
name: engineered_foreign_transposable_element_gene
intersection_of: SO:0000111 ! transposable_element_gene
intersection_of: has_quality SO:0000783 ! engineered
intersection_of: has_quality SO:0000784 ! foreign

>
> c. similar to b above, delete the following SO term:
>
> [Term]
> id: SO:0000293
> name: engineered_foreign_repetitive_element
> is_a: SO:0000280 ! engineered_gene
> is_a: SO:0000292 ! repetitive_element
>
>
> since it can be accommodated by:
>
> repetitive_element ; SO:0000292
> engineered_foreign_sequence; SO:NEW
>
And this one...
> d. it then makes sense to add an is_a relationship for engineered_gene 
> ;
> SO:0000280 so that it and all its children are under 
> engineered_sequence
> as well as gene_class:
>
> [Term]
> id: SO:0000280
> name: engineered_gene
> is_a: SO:NEW ! engineered_sequence
> is_a: SO:0000009 ! gene_class
Added engineered_gene
>

[Term]
id: SO:0000280
name: engineered_gene
intersection_of: SO:0000704 ! gene
intersection_of: has_quality SO:0000783 ! engineered

>
> 3. we need one more SO term to be able to accommodate a class of 
> foreign
> sequences that they put in the fly - they have this habit of attaching
> epitope/subcellular location "tags" to Drosophila genes and we need a
> term for that type of thing:
>
> [Term]
> id: SO:NEW
> name: engineered_tag_sequence
> def: "Sequence used in transgenes to mark genes or their products. 
> Includes epitope tags\, sequence tags\, function tags (such as 
> subcellular localisation sequences\, enzymes)." [FB:gm]
> is_a: SO:NEW ! engineered_sequence
>
Added engineered_tag
[Term]
id: SO:0000807
name: engineered_tag
namespace: sequence
intersection_of: SO:0000324 ! tag
intersection_of: has_quality SO:0000783 ! engineered

>
> 4. the following 4 SO terms have been obsoleted, but we have 
> chromosomal
> aberrations of these classes in our files, so need to use the terms.
> Could they be resurrected again ?
>
I have resurrected these terms from the dead and added the definitions 
you gave me.
They have new ids.
> [Term]
> id: SO:0000437
> name: assortment_derived_duplication
> is_obsolete: true
>
> [Term]
> id: SO:0000554
> name: assortment_derived_deficiency_plus_duplication
> is_obsolete: true
>
>
> [Term]
> id: SO:0000052
> name: assortment_derived_deficiency
> synonym: "assortment-derived_deficiency" []
> is_obsolete: true
>
>
> [Term]
> id: SO:0000058
> name: assortment_derived_aneuploid
> synonym: "assortment-derived_aneuploid" []
> is_obsolete: true
>
>
>
> If you could let me know when you have made the changes, or if you
> disagree with them (!) that would be great, as I need the new terms (or
> some equivalent) before I can start adding SO terms into the FlyBase
> files, thanks,
>
> Gillian