NAL-i5K / general_issues

for issues and discussions not tied to a specific repository
2 stars 0 forks source link

SOFP database warning when migrating to tripal 3 #13

Closed bradfordcondon closed 3 years ago

bradfordcondon commented 5 years ago

I get this warning migrating to Tripal 3:

Could not find details about the vocabulary: SOFP. Note: if this vocabulary does exist, try            [status]
re-populating the db2cv_mview materialized view at Admin > Tripal > Data Storage > Chado > Materialized
views.

I see there are 43 records in dbxref to this DB (SOFP), but no record on EBI.

select DISTINCT cv.name from chado.cv cv INNER JOIN chado.cvterm cvt ON cvt.cv_id = cv.cv_id where cvt.dbxref_id in (SELECT dbx.dbxref_id from chado.dbxref dbx where dbx.db_id = 43);
feature_property

the dbxrefs are used, linked to the featureproperty cv. So, really, this should be linked in the db2cv_mview. is it?

select * from chado.db2cv_mview where db_id = 43;
cv_id   cvname  db_id   dbname  num_terms
(0 rows)

nope.

This is probably a core bug, or a half-thought out core feature where it ignores the featureprop cv. I might make a core tripal issue to prevent confusion for other sites migrating.

bradfordcondon commented 5 years ago

I'm now guessing SOFP stands for Sequence Ontology Feature Property. I can find these terms in use in the featureprop table (so far looks like just "comment" is used, though).

select DISTINCT cv.name, cvt.name from chado.cv cv INNER JOIN chado.cvterm cvt ON cvt.cv_id = cv.cv_id where cvt.dbxref_id in (SELECT dbx.dbxref_id from chado.dbxref dbx where dbx.db_id = 43);
name    name
feature_property    sp_comment
feature_property    protein_id
feature_property    validation_flag
feature_property    qseq_type
feature_property    score
feature_property    status
feature_property    missing_start_codon
feature_property    citation
feature_property    dev_stage
feature_property    reported_pr_change
feature_property    keywords
feature_property    problem
feature_property    na_change
feature_property    date
feature_property    cyto_range
feature_property    evidenceGB
feature_property    missing_stop_codon
feature_property    synonym
feature_property    anticodon
feature_property    chromosome
feature_property    comment
feature_property    readthrough_stop_codon
feature_property    unixdate
feature_property    finished
feature_property    pr_change
feature_property    ontology_term
feature_property    linked_to
feature_property    map
feature_property    reported_na_change
feature_property    mol_type
feature_property    source
feature_property    description
feature_property    sp_status
feature_property    feature_property
feature_property    Gap
feature_property    encoded_symbol
feature_property    symbol
feature_property    orf_classification
feature_property    organism
feature_property    owner
feature_property    internal_synonym
feature_property    gbunit
feature_property    dicistronic
feature_property    element
feature_property    non_canonical_start_codon
feature_property    Note
feature_property    aminoacid
(47 rows)

SOFP are, indeed, SO terms- see for example https://www.ebi.ac.uk/ols/search?q=non_canonical_start_codon

So, i think that the DBs for these terms need to actually be switched from SOFP to SO .

Hmmm on the other hand, some terms are NOT in OLS. for example, "aminoacid" or ""orf_classification".

Are these custom ontology terms? if so no problem we just want to know so we can fix the (harmless) warning.

bradfordcondon commented 5 years ago

this issue has been reported by another user which means its very likely either a problem introduced by T2 or by an old version of the sequence ontology: https://github.com/tripal/tripal/issues/892

bradfordcondon commented 5 years ago

resolved with this PR https://github.com/tripal/tripal/pull/913

mpoelchau commented 4 years ago

@Ferrisx4 can you check whether this is resolved?

mpoelchau commented 4 years ago

Oh wait, looks like I found this recently. It seems to happen after every drush cache clear. Do we use this PR? https://github.com/tripal/tripal/pull/913

Ferrisx4 commented 4 years ago

Yes, that PR was made on 7.x-3.x, and is in the git log:

Merge: f7ad8703 1005a548
Author: Stephen Ficklin <spficklin@gmail.com>
Date:   Tue Apr 30 14:11:06 2019 -0700

    Merge pull request #913 from tripal/892_SOFP_issue

    resolve SOFP vocabulary warning on T2 migrated sites
Ferrisx4 commented 3 years ago

I haven't seen this error in quite a while, willing to close for now.