Handle duplicate genes

[JakeSAI]

Both the large and medium datasets on /allen contain duplicate gene names with non-duplicate gene values. Assuming these samples aren't outliers, we need a process for de-duplicating gene names in the pipeline as the duplicate ids are not only obfuscating for users but cause projection to crash.

Tasks:

• Check if scanpy has any preprocessing data validation methods that may apply here
• Decide on deduplication or duplicate rejection solution with team
• Implement solution

Validation:

• Run pipeline on dataset with duplicate genes and establish that duplicate genes are handled without manual intervention