Investigate "Unknown" Differences Between Python Results and Package Results

[karissawhiting]

Patrick discovered some annotations were Unknown/NA in Python results, but had a resulting non-NA annotation when run through the oncoKB package. These were run using the protein change endpoint which requires the following columns:

Main Parameters:

- Hugo Symbol / Entrez Gene ID 
- Protein Change: Example: V600E 
	- API Data Name: proteinChange (hgv_sp_short) 
- Consequence: Annotated manually in python annotator and R annotator using - internal data 
- Protein Start: Example: 600 
	- API Data Name: proteinPosStart (protein_pos_start) 
- Protein End: Example: 600 
	- API Data Name: proteinPosEnd (protein_pos_end) 

Additional Parameters:
- Reference genome: GRCh37 (default) or GRCh38
- OncoTree Tumor Type (optional)
- Evidence Type (optional)

Note: You can manually look these up individually here: https://www.oncokb.org/ to help check.