auto-generate isolate field in BioSample metadata TSV for Flu (#3)

* add 4 flu_biosample_metadata optional fields required for making isolate output column

* added boolean and code to automatically create the "isolate" column for Flu samples when input terra data table does not include strain or isolate

* delete comment

* bump version to 1.0.6

* update AUTHORS file; update dockerfile and readme to list 1.0.6. Dockerfile will not build successfully until github release is created

AUTHORS

@@ -1 +1,2 @@
-Sage M. Wright
+Sage M. Wright
+Curtis J. Kapsak

Dockerfile

@@ -1,4 +1,4 @@
-ARG MERCURY_VER="1.0.5"
+ARG MERCURY_VER="1.0.6"
 
 FROM google/cloud-sdk:480.0.0-slim
 

README.md

@@ -27,16 +27,16 @@ Default databases by organism:
 We highly recommend using the following Docker image to run Mercury:
 
 ```bash
-docker pull us-docker.pkg.dev/general-theiagen/theiagen/mercury:1.0.5
+docker pull us-docker.pkg.dev/general-theiagen/theiagen/mercury:1.0.6
 ```
 
 The entrypoint for this Docker image is the Mercury help message. To run this container interactively, use the following command:
 
 ```bash
-docker run -it --entrypoint=/bin/bash us-docker.pkg.dev/general-theiagen/theiagen/mercury:1.0.5
+docker run -it --entrypoint=/bin/bash us-docker.pkg.dev/general-theiagen/theiagen/mercury:1.0.6
 # Once inside the container interactively, you can run the tbp-parser tool
 python3 /mercury/mercury/mercury.py -v
-# v1.0.5
+# v1.0.6
 ```
 
 ### Locally with Python

mercury/Metadata.py

@@ -30,7 +30,7 @@ def mpox_gisaid_metadata(self):
 
   def flu_biosample_metadata(self):
     biosample_required = ["submission_id", "organism", "collected_by", "collection_date", "geo_loc_name", "host", "host_disease", "isolation_source", "lat_lon"]
-    biosample_optional = ["sample_title", "isolation_type", "bioproject_accession", "attribute_package", "strain", "isolate", "culture_collection", "genotype", "host_age", "host_description", "host_disease_outcome", "host_disease_stage", "host_health_state", "host_sex", "host_subject_id", "host_tissue_sampled", "passage_history", "pathotype", "serotype", "serovar", "specimen_voucher", "subgroup", "subtype", "description"] 
+    biosample_optional = ["sample_title", "isolation_type", "bioproject_accession", "attribute_package", "strain", "isolate", "culture_collection", "genotype", "host_age", "host_description", "host_disease_outcome", "host_disease_stage", "host_health_state", "host_sex", "host_subject_id", "host_tissue_sampled", "passage_history", "pathotype", "serotype", "serovar", "specimen_voucher", "subgroup", "subtype", "description", "abricate_flu_type" , "abricate_flu_subtype", "state", "year"] 
     return biosample_required, biosample_optional
 
   def bankit_metadata(self):

mercury/Table.py

@@ -179,6 +179,12 @@ def make_biosample_csv(self):
       else:
         biosample_metadata[column] = ""
 
+    # if either "isolate" or "strain" columns do not exist in input terra table, then set a boolean variable "user_supplied_isolate_or_strain" to False; otherwise set to True
+    if "isolate" not in self.table.columns and "strain" not in self.table.columns:
+      user_supplied_isolate_or_strain = False
+    else:
+      user_supplied_isolate_or_strain = True
+
     biosample_metadata.rename(columns={"submission_id" : "sample_name"}, inplace=True)
 
     if self.organism == "mpox" or self.organism == "sars-cov-2":
@@ -188,7 +194,15 @@ def make_biosample_csv(self):
       biosample_metadata.rename(columns={"collecting_lab" : "collected_by", "host_sci_name" : "host", "patient_gender" : "host_sex", "patient_age" : "host_age"}, inplace=True)      
       if self.organism == "sars-cov-2":
         biosample_metadata.rename(columns={"treatment" : "antiviral_treatment_agent"}, inplace=True)
-
+    # Flu only: when user does not supply isolate or strain metadata columns, create "isolate" column using the syntax below
+    elif self.organism == "flu" and user_supplied_isolate_or_strain == False :
+      # type/state/submission_id/year (subtype)
+      # strip off "Type_" from beginning of Type, e.g. "Type_A" -> "A"
+      self.logger.debug("DEBUG:User did not supply isolate or strain metadata columns, creating isolate column for Flu samples now...")
+      biosample_metadata["isolate"] = (biosample_metadata["abricate_flu_type"].str.replace("Type_","") + "/" + biosample_metadata["state"] + "/" + biosample_metadata["sample_name"] + "/" + biosample_metadata["year"] + " (" + biosample_metadata["abricate_flu_subtype"] + ")")
+    # Remove 4 extra columns from the output table prior to creating TSV file (these are simply used to create the isolate column)
+    biosample_metadata.drop(["abricate_flu_type", "abricate_flu_subtype", "year", "state"], axis=1, inplace=True)
+
     biosample_metadata.to_csv(self.output_prefix + "_biosample_metadata.tsv", sep='\t', index=False)
     self.logger.debug("TABLE:BioSample metadata file created")
 

mercury/__init__.py

@@ -1 +1 @@
-__VERSION__ = "v1.0.5"
+__VERSION__ = "v1.0.6"