Merge pull request #758 from robinschmid/food_masst_tree

[foodmasst] argument for local file

search_single_spectrum/search_single_spectrum/tool.xml

@@ -152,6 +152,7 @@
     <pathSet base="$base">
         <pathVar name="foodomics_track.script" path="foodomics_track.py" />
         <pathVar name="foodomics_metadata.file" path="gfop_ontology_foodmasst.txt" />
+        <pathVar name="foodomics_metadata_table.file" path="foodomics_metadata_foodmasst.tsv" />
     </pathSet>
     <tool name="foodomics_track">
         <require name="all_dataset_file_matches" type="file"/>
@@ -161,6 +162,7 @@
         <execution env="binary" argConvention="adhoc">
             <arg pathRef="foodomics_track.script"/>
             <arg pathRef="foodomics_metadata.file"/>
+            <arg pathRef="foodomics_metadata_table.file"/>
             <arg valueRef="all_dataset_file_matches"/>
             <arg valueRef="foodomics_tracking_file"/>
             <arg valueRef="filtered_food_metadata_file"/>

search_single_spectrum/test/test_foodomics.py

@@ -15,12 +15,14 @@ class TestLoaders(unittest.TestCase):
     def test(self):
         input_filename = "data/foodomics/all_dataset_matchs.tsv"
         foodomics_metadata = "../tools/search_single_spectrum/gfop_ontology_foodmasst.txt"
+        foodomics_metadata_table = "../tools/search_single_spectrum/foodomics_metadata_foodmasst.tsv"
 
         out_foodmasst = "tmp/out_foodmasst.tsv"
         out_filtered_food_metadata = "tmp/out_filtered_food_metadata.tsv"
 
         # run
-        foodomics_track.combine_food_masst(foodomics_metadata, input_filename, out_foodmasst, out_filtered_food_metadata)
+        foodomics_track.combine_food_masst(foodomics_metadata, foodomics_metadata_table, input_filename, out_foodmasst, \
+        out_filtered_food_metadata)
 
         # ground truth test file
         test_output = pd.read_csv("data/foodomics/foodmasst_test_output.tsv", sep="\t")
@@ -35,6 +37,27 @@ def test(self):
         pd.testing.assert_frame_equal(test_output, enrichment_df, check_like=True)
 
 
+    def test_empty_matches(self):
+        input_filename = "data/foodomics/all_dataset_matchs_empty.tsv"
+        foodomics_metadata = "../tools/search_single_spectrum/gfop_ontology_foodmasst.txt"
+        foodomics_metadata_table = "../tools/search_single_spectrum/foodomics_metadata_foodmasst.tsv"
+
+        out_foodmasst = "tmp/out_foodmasst_empty.tsv"
+        out_filtered_food_metadata = "tmp/out_filtered_food_metadata_empty.tsv"
+
+        # run
+        foodomics_track.combine_food_masst(foodomics_metadata, foodomics_metadata_table, input_filename, out_foodmasst, \
+        out_filtered_food_metadata)
+
+        # should be empty file
+        self.assertTrue(os.path.isfile(out_foodmasst), "No enrichment file exported")
+        self.assertFalse(os.path.getsize(out_foodmasst) > 0, "Enrichment file is not empty - was expected for an empty "
+                                                             "input")
+
+        self.assertTrue(os.path.isfile(out_filtered_food_metadata), "No metadata table file exported")
+        self.assertFalse(os.path.getsize(out_filtered_food_metadata) > 0, "Metadata table file is not empty - was "
+                                                                          "expected for an empty input")
+
     def test_food_tree(self):
         test_output = "data/foodomics/foodmasst_test_output.tsv"
         in_html = "../tools/search_single_spectrum/interactive_tree_js/collapsible_tree_v3.html"

search_single_spectrum/tools/search_single_spectrum/foodomics_track.py

@@ -10,6 +10,10 @@
 import pandas as pd
 import os
 import argparse
+import logging
+
+logging.basicConfig(level=logging.DEBUG)
+logger = logging.getLogger(__name__)
 
 
 def calculate_enrichment(matches_df, metadata_df):
@@ -48,10 +52,11 @@ def calculate_enrichment(matches_df, metadata_df):
     return pd.DataFrame(output_list)
 
 
-def metadata_file_matches(matches_df):
+def metadata_file_matches(matches_df, foodomics_metadata_table):
     """
     Create filtered metadata table for MASST matched files
     :param matches_df: filtered MASST matches in GFOP files : pandas dataframe
+    :param foodomics_metadata_table: filtered metadata table file
     :return: filtered metadata
     """
 
@@ -60,7 +65,8 @@ def metadata_file_matches(matches_df):
     # /foodomics_metadata_foodmasst.txt', sep='\t')
     # gfop_meta = pd.read_csv(
     #     'https://github.com/global-foodomics/GFOPontology/raw/master/data/foodomics_metadata_foodmasst.tsv', sep='\t')
-    gfop_meta = pd.read_csv('foodomics_metadata_foodmasst.tsv', sep='\t')
+    # gfop_meta = pd.read_csv('foodomics_metadata_foodmasst.tsv', sep='\t')
+    gfop_meta = pd.read_csv(foodomics_metadata_table, sep='\t')
 
     # Removing extensions on both
     gfop_meta["filename"] = gfop_meta["filename"].apply(lambda x: os.path.splitext(x)[0])
@@ -80,7 +86,8 @@ def filter_matches(matches_df):
     return matches_df[matches_df["dataset_id"] == "MSV000084900"]
 
 
-def combine_food_masst(foodomics_metadata, matches_results, output_enrichment, output_metadata_matches):
+def combine_food_masst(foodomics_metadata, foodomics_metadata_table, matches_results, output_enrichment,
+                       output_metadata_matches):
     try:
         matches_df = pd.read_csv(matches_results, sep="\t")
         # filter to global foodomics
@@ -93,30 +100,40 @@ def combine_food_masst(foodomics_metadata, matches_results, output_enrichment, o
             sorted_df = enrichment_df.sort_values(by=["group_value"])
             sorted_df.to_csv(output_enrichment, sep="\t", index=False)
 
-        matched_metadata = metadata_file_matches(matches_df)
+        # a table of matched files and their metadata (filtered columns)
+        matched_metadata = metadata_file_matches(matches_df, foodomics_metadata_table)
         if output_metadata_matches is not None:
             matched_metadata.to_csv(output_metadata_matches, sep="\t", index=False)
 
         return enrichment_df, matched_metadata
-    except:
+    except Exception as e:
+        # exit with error
+        logger.exception(e)
+        logger.exception("Error during foodMASST creation of output")
+
         # on error write all files as empty
-        with open(output_enrichment, "w") as o:
+        with open(output_enrichment, "a") as o:
             pass
             # o.write("")
-        with open(output_metadata_matches, "w") as o:
+        with open(output_metadata_matches, "a") as o:
             pass
             # o.write("")
 
 
 def main():
     parser = argparse.ArgumentParser(description='Create foodomics enrichment')
-    parser.add_argument('foodomics_metadata', help='the foodomics metadata table')
+    # foodomics_metadata maps the raw data files into all matching ontology classes
+    # foodomics_metadata_table is a filtered version with extensive metadata for each file
+    parser.add_argument('foodomics_metadata', help='the foodomics ontology table (data file and all ontology levels)')
+    parser.add_argument('foodomics_metadata_table', help='the foodomics metadata table (data file and filtered '
+                                                         'metadata)', default='foodomics_metadata_foodmasst.tsv')
     parser.add_argument('matches_results', help='MASST match results')
     parser.add_argument('output_enrichment', help='Output tsv file for enrichment results')
     parser.add_argument('metadata_matches', help='Output filtered metadata tsv file')
     args = parser.parse_args()
 
-    combine_food_masst(args.foodomics_metadata, args.matches_results, args.output_enrichment, args.metadata_matches)
+    combine_food_masst(args.foodomics_metadata, args.foodomics_metadata_table, args.matches_results,
+                       args.output_enrichment, args.metadata_matches)
 
 
 if __name__ == "__main__":