#230 stats : reports: relations used

[Valerie Wood]

Could you update this list for us at some point?
28 | condition
1 | dependent_on
179 | during
3 | exists_during
119 | has_part
10 | has_regulation_target
113 | has_substrate
4 | independent_of
20 | localization_dependent_on
998 | localizes
8 | modified_by
5 | occurs_in
156 | part_of
16 | requires_feature
1 | requires_regulator

I wondered if I could access this within the curations tool.

I found the list of relations
http://oliver0.sysbiol.cam.ac.uk/pombe/view/object/cv/6?model=chado
but they are all "0"

[Kim Rutherford]

The list at http://oliver0.sysbiol.cam.ac.uk/pombe/view/object/cv/6?model=chado only shows the number of times a term is used to annotate a feature. These are relations rather than terms so the number of times they are used won't be included in the count. I'll send the updated count list once the sysbiol fileserver comes back to life - it's causing problems for the pombe server.

[Kim Rutherford]

How does this look? The query to make this table is quite hairy, so it may be wrong.

We should be able to have this as a report in the curation tool. I've moved this to the curation tool tracker and I'll leave it open.

select count(id), name from (select p.cvterm_id::text || '_cvterm' as id, substring(type.name from 'annotation_extension_relation-(.*)') as name from cvterm type, cvtermprop p where p.type_id = type.cvterm_id and type.name like 'annotation_ex%' union all select r.cvterm_relationship_id::text || '_cvterm_rel' as id, t.name as name from cvterm_relationship r, cvterm t where t.cvterm_id = type_id and r.subject_id in (select cvterm_id from cvterm, cv where cvterm.cv_id = cv.cv_id and cv.name = 'PomBase annotation extension terms')) as sub where name <> 'is_a' group by name order by name;

count | name
-------+---------------------------
1 | dependent_on
6 | during
115 | exists_during
58 | happens_during
1 | has_downstream_target
82 | has_regulation_target
105 | has_substrate
2 | independent_of
20 | localization_dependent_on
138 | localizes
9 | modified_by
1 | occurs_in
17 | requires_feature
1 | requires_localization
1 | requires_regulator
4 | unfolds_around

[Valerie Wood]

This does look ball park right for most.
There may be a small problem with "localizes"

138 | localizes

In the flat files we have
grep "localizes" *.contig | wc
89
PLUS
"localization_dependency" *.contig |wc
192
(which were mapped to localizes during import)

This should give ups a minimum of 281....
are we "losing" some somewhere?

[Kim Rutherford]

I'll look into the localizes count.