feat: add tissue_type and new validation rules for tissue_ontology_term_id and cell_type_ontology_term_id #623
Conversation
…ue_ontology_term_id and cell_type_ontology_term_id
Codecov Report
@@ Coverage Diff @@
## main #623 +/- ##
==========================================
- Coverage 83.08% 83.01% -0.08%
==========================================
Files 19 19
Lines 1709 1684 -25
==========================================
- Hits 1420 1398 -22
+ Misses 289 286 -3
Flags with carried forward coverage won't be shown. Click here to find out more.
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
| @@ -281,16 +281,31 @@ def test_assay_ontology_term_id(self): | |||
|
|
|||
| def test_cell_type_ontology_term_id(self): | |||
| """ | |||
| cell_type_ontology_term_id categorical with str categories. This MUST be a CL term. | |||
| cell_type_ontology_term_id categorical with str categories. This MUST be a CL term, and must NOT match forbidden | |||
There was a problem hiding this comment.
Do we have a locally glossary somewhere for these acronyms(CL)?
There was a problem hiding this comment.
I don't think its an acronym, its a prefix used for cell ontology terms. There's a glossary in the user-facing schema documentation
| ) | ||
| with self.subTest(forbidden_term="EFO:0000001"): | ||
| self.validator.adata.obs.loc[ | ||
| self.validator.adata.obs.index[0], "cell_type_ontology_term_id" |
There was a problem hiding this comment.
What is the benefit of using the numpy method of slicing foo[ x, y], as opposed to pythons way using foo[x][y]?
There was a problem hiding this comment.
this is the syntax for the loc accessor function for pandas dataframes
There was a problem hiding this comment.
I believe this is the standard way to reassign a pandas df entry but perhaps both ways work
#514
#517
Changes:
- do NOT accept suffixes in term_ids and do not append them to the resulting label
- enforce cell_type_ontology_term_id rules on this term_id IF tissue_type is 'cell culture',
- enforce term_id is child term of UBERON:0001062 if tissue_type is 'organoid' or 'tissue'