✨ add multiple phenotypes as a list

src/data/import_multiple_phenotypes.py

@@ -15,11 +15,34 @@
 from click_option_group import optgroup, RequiredMutuallyExclusiveOptionGroup
 from pathlib import Path
 
-from src.features.smarterdb import global_connection
+from src.features.smarterdb import global_connection, Phenotype
+from src.data.common import pandas_open, deal_with_datasets, get_sample_species
+from src.features.utils import sanitize
 
 logger = logging.getLogger(__name__)
 
 
+def create_or_update_phenotype(
+        sample, phenotype: dict):
+
+    if not phenotype:
+        logger.debug(f"Skipping {sample}: nothing to update")
+        return
+
+    if not sample.phenotype:
+        logger.debug(f"Create a new phenotype for {sample}")
+        sample.phenotype = Phenotype()
+
+    for key, value in phenotype.items():
+        setattr(sample.phenotype, key, value)
+
+    logger.info(
+        f"Updating '{sample}' phenotype with '{sample.phenotype}'")
+
+    # update sample
+    sample.save()
+
+
 @click.command()
 @click.option(
     '--src_dataset', type=str, required=True,
@@ -43,14 +66,15 @@
 @optgroup.option('--alias_column', type=str, help="An alias for original_id")
 @click.option(
     '--column',
+    'columns',
     required=True,
     multiple=True,
     help=(
         "Column to track. Could be specified multiple times")
 )
 @click.option('--na_values', type=str, help="pandas NA values")
 def main(src_dataset, dst_dataset, datafile, sheet_name, breed_column,
-         id_column, alias_column, column, na_values):
+         id_column, alias_column, columns, na_values):
     """Read multiple data for the same sample from phenotype file and add it
     to SMARTER-database samples"""
 
@@ -61,7 +85,35 @@ def main(src_dataset, dst_dataset, datafile, sheet_name, breed_column,
             "Loading multiple phenotypes by breed or alias is not yet "
             "implemented")
 
-    logger.debug(f"Reading {column} columns")
+    logger.debug(f"Reading {columns} columns")
+
+    src_dataset, dst_dataset, datapath = deal_with_datasets(
+        src_dataset, dst_dataset, datafile)
+
+    SampleSpecie = get_sample_species(dst_dataset.species)
+
+    if sheet_name and sheet_name.isnumeric():
+        sheet_name = int(sheet_name)
+
+    # open data with pandas
+    data = pandas_open(datapath, na_values=na_values, sheet_name=sheet_name)
+
+    # process unique ids
+    for id_ in data[id_column].unique():
+        subset = data[data[id_column] == id_]
+
+        phenotype = {}
+
+        for column in columns:
+            phenotype[sanitize(column)] = subset[column].to_list()
+
+        original_id = str(id_)
+
+        # ok iterate over all samples of this dataset
+        for sample in SampleSpecie.objects.filter(
+                dataset=dst_dataset, original_id=original_id):
+
+            create_or_update_phenotype(sample, phenotype)
 
     logger.info(f"{Path(__file__).name} ended")
 

tests/data/test_import_multiple_phenotypes.py

@@ -53,9 +53,12 @@ def setUpClass(cls):
         cls.sheet.cell(row=2, column=1, value="test-1")
         cls.sheet.cell(row=2, column=2, value=123)
         cls.sheet.cell(row=2, column=3, value=11.5)
-        cls.sheet.cell(row=2, column=1, value="test-1")
-        cls.sheet.cell(row=2, column=2, value=456)
-        cls.sheet.cell(row=2, column=3, value=23.0)
+        cls.sheet.cell(row=3, column=1, value="test-1")
+        cls.sheet.cell(row=3, column=2, value=456)
+        cls.sheet.cell(row=3, column=3, value=23.0)
+        cls.sheet.cell(row=4, column=1, value="test-2")
+        cls.sheet.cell(row=4, column=2, value=123)
+        cls.sheet.cell(row=4, column=3, value=11.5)
 
     @classmethod
     def tearDownClass(cls):
@@ -97,7 +100,7 @@ def test_help(self):
 class TestImportPhenotypeBySamples(PhenotypeMixin, unittest.TestCase):
     @patch('src.features.smarterdb.Dataset.working_dir',
            new_callable=PropertyMock)
-    def test_import_phenotype(self, my_working_dir):
+    def test_import_multiple_phenotype(self, my_working_dir):
         # create a temporary directory using the context manager
         with tempfile.TemporaryDirectory() as tmpdirname:
             working_dir = pathlib.Path(tmpdirname)