Mutation lollipop view hangs when gene has unknown Entrez. This could happen for "negative Entrez scenario".
Steps to reproduce on cbioportal.org (and on /beta):
CDKN2AP16INK4A . Link: http://www.cbioportal.org/index.do?Action=Submit&genetic_profile_ids=msk_impact_2017_mutations&case_set_id=msk_impact_2017_all&cancer_study_id=msk_impact_2017&gene_list=CDKN2AP16INK4A&tab_index=tab_visualize&#mutation_details&show_samples=false&Details:

Desired solution:
@jjgao also happens on /beta btw.
@oplantalech, @alisman: the CIVIC icon also hangs, but perhaps this is because https://github.com/cBioPortal/cbioportal-frontend/pull/624 is not deployed yet.
@jjgao, @alisman : as a general comment for all external services we are integrating: I would recommend reviewing the exception handling scenarios in all cases. Any of these services might fail at any time (including the newly integrated g2s.genomenexus.org). Ideally the UIs won't hang when this happens but give a clear indication about the temporary error in such a case.
@pieterlukasse agreed. we have a global error handler that shows a dialog. we disabled it a couple weeks ago because it was bring down the patient view unnecessarily. but we need to bring it back with a more specific treatment of different errors which may or may not be "catastrophic"
@onursumer lets discuss how to handle this in UI. Also, we need a way to egage the global error handler and define an error message for each error (so the dialog can display)
@alisman here is my desired solution for this specific case:
@pieterlukasse @alisman Soon we are going to update the way we obtain the domain information: We are switching from pfam/graphics API to genome nexus API. It might be a good idea to deal with this issue while redesigning/reimplementing domain rendering.
With the new API, we won't need to query for uniprot id with entrez gene id, because we won't need uniprot id for pfam graphics data any more. But, there will still be unknown genes, i.e. genes with no protein (length) information. And we will still need the gene length / 3 approach for them.
@pieterlukasse I am just trying to patch a quick fix for now by using the gene length, but looks like we don't have that information in our database for the gene CDKN2AP16INK4A. Do you remember how we were getting the gene length previously?
@pieterlukasse My bad, actually we have the length information. Looks like currently we only fetch the minimum info (i.e: projection=ID), that's why I was not able to see the length in the result.