Merge pull request #126 from phac-nml/fix/keyerror

Fixes staramr so it works for newer pandas

.travis.yml

@@ -1,8 +1,8 @@
 sudo: required
 language: python
 python:
-  - "3.5"
   - "3.6"
+  - "3.7"
 
 env:
   matrix:

CHANGELOG.md

@@ -1,3 +1,7 @@
+# Version 0.7.2
+
+* Fixed `KeyError` issue with later versions of pandas (#115, thanks @javiertognarelli).
+
 # Version 0.7.1
 
 * Fix a bug so that the Sequence column in resfinder.tsv uses the isolate sequence instead of the reference sequence

README.md

@@ -232,6 +232,8 @@ staramr db restore-default
 * Git
 * MLST
 
+**Note: if you wish to use Python 3.5 you will have to install BioPython <= 1.76 as later versions no longer support Python 3.5.**
+
 # Input
 
 ## List of genes to exclude

staramr/__init__.py

@@ -1 +1 @@
-__version__ = '0.7.1'
+__version__ = '0.7.2'

staramr/results/AMRDetectionSummary.py

@@ -37,12 +37,12 @@ def __init__(self, files, resfinder_dataframe: DataFrame, quality_module_datafra
 
     def _compile_results(self, resistance_frame: DataFrame) -> DataFrame:
         df_summary = resistance_frame.sort_values(by=['Gene']).groupby(['Isolate ID']).aggregate(
-            lambda x: {'Gene': (self.SEPARATOR + ' ').join(x['Gene'])})
+            lambda x: {'Gene': (self.SEPARATOR + ' ').join(x.get('Gene'))})
         return df_summary[['Gene']]
 
     def _compile_plasmids(self, plasmid_frame: DataFrame) -> DataFrame:
         ds_summary = plasmid_frame.sort_values(by=['Gene']).groupby(['Isolate ID']).aggregate(
-            lambda x: {'Gene': (self.SEPARATOR + ' ').join(x['Gene'])})
+            lambda x: {'Gene': (self.SEPARATOR + ' ').join(x.get('Gene'))})
 
         ds_frame = ds_summary[['Gene']]
 

staramr/results/AMRDetectionSummaryResistance.py

@@ -22,7 +22,7 @@ def __init__(self, files, resfinder_dataframe, quality_module_dataframe,pointfin
         super().__init__(files, resfinder_dataframe,quality_module_dataframe, pointfinder_dataframe, plasmidfinder_dataframe, mlst_dataframe)
 
     def _aggregate_gene_phenotype(self, dataframe):
-        flattened_phenotype_list = [y.strip() for x in dataframe['Predicted Phenotype'].tolist() for y in
+        flattened_phenotype_list = [y.strip() for x in dataframe.get('Predicted Phenotype').tolist() for y in
                                     x.split(self.SEPARATOR)]
         uniq_phenotype = OrderedDict.fromkeys(flattened_phenotype_list)