Synchronize cohorts with the NHGRI GWAS Catalog, when available #394
Conversation
ens-lgil
commented
Oct 17, 2024
ens-lgil
commented
- During the data import
- During the release process (updating GWAS Catalog entries)
| @@ -37,6 +35,26 @@ def get_gwas_info(self,sample): | |||
| if response_data: | |||
| try: | |||
| source_PMID = response_data['publicationInfo']['pubmedId'] | |||
|
|
|||
| # Create list of cohorts if it exists in the GWAS study | |||
| # This override the Cohorts found previously in the cohort column in the spreadsheet | |||
There was a problem hiding this comment.
Can we get a warning/message if the spreadsheet cohorts are replaced? (including previous and new values)
There was a problem hiding this comment.
Ah yes, we can add that
There was a problem hiding this comment.
It will look like that:
# GCST00XXXX:
1 distinct ancestries
{'source_PMID': '1234567', 'sample_number': 23464, 'ancestry_broad': 'European', 'ancestry_country': 'U.K.'}
/!\ Replacing cohorts list:
- Old set: ACTS, LASA
+ New set: UKB
>> SCORE updated: PGS00XXXX
with the new bits:
/!\ Replacing cohorts list:
- Old set: ACTS, LASA
+ New set: UKB
There was a problem hiding this comment.
It might useful to know if it's just additions or things are missing? It's more likley to be correct if it's adding more annotations, but if it's removing an annotation we should be careful.
There was a problem hiding this comment.
If it's using the GCST, it should match what is in the GWAS Catalog ?
At the moment this script overwrite the Sample if the sample_number, sample_cases and sample_controls are NULL, i.e. GCSTs which were not released when the study was imported.
…he one alreday stored in the database
… from the GWAS study (fetched via the GWAS REST API)
smlmbrt
changed the title
smlmbrt
changed the title
…AS study) is merged with the list of cohorts from the GWAS Catalog REST API
