knowledgesystems / cmo-pipelines

GNU Affero General Public License v3.0
3 stars 10 forks source link

Issue merging Archer fusions in Archer only cohort? #716

Closed n1zea144 closed 4 years ago

n1zea144 commented 4 years ago

Here is a patient in MSK-IMPACT that has archer fusions that are properly displayed:

mskimpact-cohort

but in the Archer only cohort an error is thrown. Do we properly merge fusion data in the Archer only cohort into the mutation profile?

archer-cohort

I noticed that sometimes there is an error message:

image

and sometimes there is an empty patient view:

image

ao508 commented 4 years ago

Created an issue in the frontend https://github.com/cBioPortal/cbioportal-frontend/issues/3275

Shared direct links to the same patient's patient view page in the gdac mskimpact study and the triage mskarcher_test study with Onur.

ao508 commented 4 years ago

Confirmed with Onur that this is a frontend issue with how variants are filtered/displayed on frontend in mutation table.

ao508 commented 4 years ago

At this time there is nothing else for us to do - when the fix is in place we will need to update the meta_fusions.txt file for the archer only cohort. (stable id, genetic alteration type need updating)

ao508 commented 4 years ago

Ritika shared that genes with fusions table on study summary page does not render for ARCHER only cohort.

ao508 commented 4 years ago

There's an ongoing conversation in the issue linked above regarding the proper way to handle the ARCHER only cohort.

ao508 commented 4 years ago

@n1zea144 I wanted to make a quick note - the patient linked above does not have any ARCHER fusion events, just fusion events reported by the IMPACT pipeline.

ao508 commented 4 years ago

Here is a patient that has ARCHER events:

https://cbioportal.mskcc.org/patient?studyId=mskimpact&caseId=P-0000104#navCaseIds=mskimpact:P-0000104

https://cbioportal.mskcc.org/patient?studyId=mskarcher&caseId=P-0000104#navCaseIds=mskarcher:P-0000104

@n1zea144 The ARCHER fusion event displays in both cohorts for this patient so I'm not sure if there is an error to address after all?

The example patient that I was using as a reference above does not have ARCHER fusion events so that led me to believe that the frontend was incorrectly omitting the fusion records from the patient view mutation table.