feat: possibility to manually exclude metadata columns (#67)

.tests/config/config.yaml

@@ -14,6 +14,8 @@ input: ../incoming/
 data: data/
 # Path to the sheet holding metadata information
 metadata: config/pep/metadata.txt
+# Remove columns from analysis
+remove-columns: ["site-name"]
 # Paths to the databases used for classification and taxonomy
 database:
   download-path-seq: resources/silva-138-99-seqs.qza

config/config.yaml

@@ -14,6 +14,8 @@ input: ../incoming/
 data: data/
 # Path to the sheet holding metadata information
 metadata: config/pep/metadata.txt
+# Remove columns from analysis
+remove-columns: ["site-name"]
 # Paths to the databases used for classification and taxonomy
 database:
   download-path-seq: https://data.qiime2.org/2022.2/common/silva-138-99-seqs.qza

workflow/rules/common.smk

@@ -90,6 +90,11 @@ def get_reads_for_kraken():
 def get_metadata_columns():
     metadata = pd.read_csv(config["metadata"], header=0, delimiter=",")
     header = metadata.columns[metadata.isin(["numeric"]).any()].values.tolist()
+    remove_list = config["remove-columns"]
+    if len(remove_list) != 0:
+        for item in remove_list:
+            if item in header:
+                header.remove(item)
     return header
 
 
@@ -106,5 +111,10 @@ def get_metadata_categorical_columns():
         if "barcode" in name:
             namelist.remove(name)
         y = y + 1
+    remove_list = config["remove-columns"]
+    if len(remove_list) != 0:
+        for item in remove_list:
+            if item in header:
+                namelist.remove(item)
     print(namelist)
     return namelist