Use drug class instead of gene name to query for high level class #278
Conversation
…high level drug class values
workflows/amr/run.wdl
Outdated
| if 'highLevelDrugClasses' not in ontology[gene_name]: | ||
| return [] | ||
| return ontology[gene_name]['highLevelDrugClasses'] | ||
| def get_high_level_classes(drug_classes_union): |
There was a problem hiding this comment.
what's the form of drug_classes_union? A set of strings like "rifamycin antibiotic; peptide antibiotic"? Wondering why it would have length > 1 ever
There was a problem hiding this comment.
It's type is list[str] but there is usually only one element in the list, a string containing drug classes separated by semicolons, e.g.:
The variable is created by this line of code:
dc = remove_na(set(sub_df['Drug Class_contig_amr']).union(set(sub_df['Drug Class_kma_amr'])))This can result in two elements in the list if the two columns differ.
There was a problem hiding this comment.
I'm hoping that soon I can move the python code out to a separate file and upgrade the container's python version to >= 3.9 to make type hinting a lot easier to add.
There was a problem hiding this comment.
In the meantime that's probably a misleading variable name, I'll change it to something better
There was a problem hiding this comment.
Ahh that makes more sense that it would have length > 1 because of two sources. Thanks for explaining!
Obtains the names of high-level drug classes for each gene by using the gene's drug class to query the ontology file.
Sample output:
