fixed header for gene_expression master table

Author: ymahlich

scripts/prepare_data_for_improve.py

@@ -250,16 +250,18 @@ def process_datasets(args):
     # moving ensemble_id & gene_symbol columns to the front of the table
     # such that when transposing the DataFrame they are row 3 and 2
     # respectively
-    merged_transcriptomics.insert(
-        1,
-        'ensembl_gene_id',
-        merged_transcriptomics.pop('ensembl_gene_id')
-    )
     merged_transcriptomics.insert(
         1,
         'gene_symbol',
         merged_transcriptomics.pop('gene_symbol')
     )
+    merged_transcriptomics.insert(
+        0,
+        'ensembl_gene_id',
+        merged_transcriptomics.pop('ensembl_gene_id')
+    )
+
+    merged_transcriptomics = merged_transcriptomics[merged_transcriptomics['entrez_id'] != 0]
 
     # writing the expression datatable to '/x_data/*_expression.tsv'
     outfile_path = args.WORKDIR.joinpath(