Re: [Gusdev-gusdev] DNA, RNA and Protein GUS Features + PeptidePropertyType Table

[Arnaud K. <ax...@sa...>]

Hi Chris

 Chris Stoeckert wrote:

> Hi Arnaud,
> I finally went through your list. These will certainly enrich GUS! 
> Some questions/issues though.
> First a general request for documentation of the tables and attributes 
> to explain what they are to be used for. We have a plug-in that takes 
> a file in the format:
>
> TableName\t\tdescription
> TableName\tAttributeName\tDescription

Sorry for the lack of documentation, I'm going to prepare a doc file.

>
> In particular, I am curious as to what InflectionPointFeature and 
> ReplicationFeature are.

In Leishmania, but more generally for any organism which has 
polycistronic transcription, the inflection point represents the start 
of the transcription. There are some studies trying to find out whether 
or not it corresponds to a conserved sequence. If so, it might 
interesting for curator to annotate them.
ReplicationFeature represents origins of replication. ReplicationFeature 
sounds more generic but they will be given a more specific SO term.

>
> For the NAFeature views you propose, are you using "source_id" to 
> point to SRes:SequenceOntology? If so, why not call the attribute "so_id"?
> Similarly, for GenomeSequence as a view of NASequence, is this what 
> "source_id" is for?

The source_id is not related to Sequence Ontology.
The main point with my proposal is to replace controlled vocabularies 
specifying the type of a feature with SO. But to do so, I think we need 
a many to many relationship between feature views and SO. Could it be 
done by using the "DoTS::GOTermAssociation 
<http://www.cbil.upenn.edu/cgi-bin/GUS30/schemaBrowser.pl?db=GUS30&table=DoTS::GOTermAssociation&path=DoTS::GOTermAssociation>" 
table or cloning it ?
As I realise it's an important point for the GUS design, please let me 
know if you agree or  if you want to propose something else.

I added a source_id to the sequence and NAfeature views because I can 
see that all feature objects have this attribute. What is this attribute 
for in GUS ?

>
> The AAFeature views have "name" attributes and I wonder whether we 
> should have a table in SRes for controlled vocabulary terms for 
> protein features that we can point to (as with sequence ontology). 
> This would avoid the uncontrolled use of "name." I notice that 
> PeptideProperty has been given a controlled vocabulary table 
> PeptidePropertyType in this regard. Rather than have a table for each, 
> we could centralize them. Any choices for the resource to use for 
> these names? SWISS-PROT?

SO doesn't cover protein features but would eventually. Anyway in the 
meantime, it makes sense to have a controlled vocabulary. I'm not aware 
of such controlled vocabulary though.

Shall I replace the PeptidePropertyType table by a more generic one, 
AAFeatureName ?

>
> Cheers,
> Chris
>
> On Tuesday, October 8, 2002, at 09:00 AM, Arnaud Kerhornou wrote:
>
>> From: Arnaud Kerhornou <ax...@sa...>
>> Date: Tue Oct 8, 2002  9:00:32 AM US/Eastern
>> To: gusdev-gusdev <gus...@li...>, 
>> gen...@li...
>> Subject: [Gusdev-gusdev] DNA, RNA and Protein GUS Features + 
>> PeptidePropertyType Table
>>
>> Hi
>>
>> I've attached the SQL statements for new views/tables in GUS3, as 
>> well as updates of existing views/tables. It covers a new sequence 
>> object and new DNA, RNA and protein features that we would like to 
>> use. Some of them have been designed to go along Sequence Ontology 
>> classification (see below).
>>
>> Here a summary of the list of new views or tables:
>>
>> Updated views/tables:
>>   * NAFeatureImp - modified:
>>      * name from varchar2(30) to varchar2(50)
>>   * RestrictionFragmentFeature - added:
>>      * type_of_cut (sticky or blunt)
>>   * SignalPeptideFeature - added:
>>      * targetting
>>   * GeneSynonym - added:
>>      * is_obsolete
>>
>> New views/tables:
>>
>> A new sequence object on the top of NASequenceImp:
>>   * GenomicSequence
>>
>> New views on the top of NAFeatureImp:
>>   * ChromosomeElement
>> NB : (centromere, telomere => SO)
>>   * InflectionPointFeature
>>   * RepeatFeature
>> NB : repeat type => SO
>>   * RepeatRegionFeature
>>   * ReplicationFeature
>>   * TransposableElementFeature
>> NB: SO would give the type
>>   * RNARegulatory
>>   * RNASecondaryStructure
>>   * SpliceSiteFeature
>>
>> New views on the top of AAFeatureImp:
>>   * AASecondaryStructure
>>   * AATertiaryStructure
>>   * DomainFeature
>>   * PeptideProperty
>>   * PostTranslationModification
>>   * TransmembraneDomainFeature
>>
>> A new table:
>>   * PeptidePropertyType
>>
>>
>> Note that the design takes into account the use of the Sequence 
>> Ontology (SO) to refine the types of some the features, eg to 
>> differentiate the different types of transposable elements, of 
>> repeats or of chromosome elements (centromere, telomere ...).
>
>
>