Dear all,
We've been using the cBioPortal API to retrieve data in the context of the Cancermuts Python package.
Recently (probably within a week's time), the data returned by the cBioPortal API has been in a different format than our package normally expects. We've been doing this to retrieve cancer studies:
...
result = self._client.Studies.getAllStudiesUsingGET().result()
...
df = pandas.DataFrame(dict([ (attr, [ getattr(entry, attr) for entry in result ]) for attr in dir(result[0]) ]))
basically, we flatten the resulting data to a dataframe, which until last week seemed to be working fine. Now, It looks like the returned data structure has changed, specifically in the cancerType column. Our cancerType column now looks something like:
TypeOfCancer(cancerTypeId=None, dedicatedColor='HotPink', id='acbc', name='Adenoid Cystic Breast Cancer', parent='brca', shortName='ACBC')
while it used to be a simple string with the cancer type name, for what I can tell.
I can easily change the parsing so that we get the right data structure and extract what we need, but I would first like to know if the change I see is intentional and there to stay for the foreseeable future, so that I can rely on it moving forward, or it's due to a mistake which is bound to be corrected eventually.
Thank you and kind regards,
____________________________
Matteo Tiberti, PhD
Staff Scientist at the Cancer Structural Biology group
Danish Cancer Society Research Center
Strandboulevarden 49
2100, Copenhagen, Denmark