adjust test folder reference

tests/integration_tests_auto_check.py

@@ -291,11 +291,13 @@
             failures += 1
 
 
+tests_dir = sys.argv[1]
+print('tests dir is {}'.format(tests_dir))
 # Single-cell vs bulk processing check for same single-cell dataset 
 sc_folder = 'only_5_cells_test'
 bulk_folder = 'only_5_cells_bulk_mode_test'
-sc_5_cells = pd.read_csv('singlecell_tests/{}/final_filtered_site_info.tsv'.format(sc_folder), sep='\t').sort_values(['position', 'strand_conversion'])
-bulk_5_cells = pd.read_csv('singlecell_tests/{}/final_filtered_site_info.tsv'.format(bulk_folder), sep='\t').sort_values(['position', 'strand_conversion'])
+sc_5_cells = pd.read_csv('{}/singlecell_tests/{}/final_filtered_site_info.tsv'.format(tests_dir, sc_folder), sep='\t').sort_values(['position', 'strand_conversion'])
+bulk_5_cells = pd.read_csv('{}/singlecell_tests/{}/final_filtered_site_info.tsv'.format(tests_dir, bulk_folder), sep='\t').sort_values(['position', 'strand_conversion'])
 print("Checking that analyzing a single-cell dataset in 'bulk' mode (i.e. not specificying the 'CB' barcode) yields the exact same positions and base changes, but with counts and coverages aggregated rather than at a single-cell resolution")
 grouped_sc = pd.DataFrame(sc_5_cells.groupby(['contig', 'position', 'strand_conversion']).agg({'count': sum, 'strand_conversion': 'unique'}))
 grouped_sc.index.names = ['contig', 'position', 'c']

tests/integration_tests_run.sh

@@ -48,7 +48,7 @@ done
 
 
 echo "Checking results..."
-$mypython $MARINE/tests/integration_tests_auto_check.py
+$mypython $MARINE/tests/integration_tests_auto_check.py $MARINE/tests
 
 exitcode=$?