In preparation for more substantial changes in the next few days, I did a PEP8/style/readability pass over Vaxrank and made changes shouldn't change functionality.
Removed unused imports
Fixed whitespace/indentation/line length
Updated copyright years
Moved FM index code into reference_proteome module, to hide implementation details from rest of logic (the calling code should only care that we're checking whether a peptide is in the reference, not we're using an FM index which returns all search results as a list).
The biggest change is in the GenePathwayCheck class, where I renamed things like _CLASS_I_MHC to _CLASS_I_MHC_COLUMN_NAME (made the code easier for me to read) and have the objects hold on to sets of the gene names and variants, rather than doing more complicated DataFrame operations for each variant.
Coverage increased (+2.2%) to 88.601% when pulling 4731cc03e2f87286be7c58dae7bd67aa786317ec on formatting-cleanup into 64dc542c3576398401b2aed82b4cde9ecdf1e67c on master.
In preparation for more substantial changes in the next few days, I did a PEP8/style/readability pass over Vaxrank and made changes shouldn't change functionality.
reference_proteome
module, to hide implementation details from rest of logic (the calling code should only care that we're checking whether a peptide is in the reference, not we're using an FM index which returns all search results as a list).The biggest change is in the
GenePathwayCheck
class, where I renamed things like_CLASS_I_MHC
to_CLASS_I_MHC_COLUMN_NAME
(made the code easier for me to read) and have the objects hold on to sets of the gene names and variants, rather than doing more complicated DataFrame operations for each variant.