Update the way to filter the list of datasets data to export, by crea…

exports/config_template.py

@@ -1,3 +1,10 @@
+## Dataset selection
+dataset_selection = { '<dataset_column>': '<dataset_value(s)>' }
+# e.g.:
+# 'publication_id': 1
+# 'num': 208
+# 'id__in': ['OPD000056','OPD000105','OPD000154','OPD000203']
+# 'num__in': ['56','105','154','203']
 
 # Metadata exports
 metadata_exports_publication_id = '<publication_id, e.g. 1>'
@@ -6,7 +13,6 @@
 # PheWAS exports
 phewas_exports_dir = '<path_to_directory>'
 
-
 # SQLite exports
 sqlite_default_values = {
     'opp_id': '<dataset_id>', # e.g. OPP000003
@@ -17,7 +23,6 @@
     'use_different_id_as_gene': '<opgs_id_OR_name>' # Use the OmicsPred ID, the score name in the 'gene' column or not. None if the key is missing
 }
 
-
 # Scoring files exports - only for PredictDB studies
 scoring_file_config = {
     'pmid': '<pubmed ID>', # e.g. 32913098

exports/datasets.py

@@ -0,0 +1,19 @@
+from django.db.models import Q
+from omicspred.models import Dataset
+from exports.config import dataset_selection
+
+class DatasetsSelection():
+
+    def __init__(self):
+        print(f'dataset_selection: {dataset_selection}')
+        ds_keys = list(dataset_selection.keys())
+        print(f'dataset_selection keys: {ds_keys}')
+        self.dataset_col = ds_keys[0]
+        self.dataset_value = dataset_selection[self.dataset_col]
+
+    def get_datasets(self) -> list:
+        param = self.dataset_col
+        param = Q(**{f'{self.dataset_col}':self.dataset_value})
+
+        datasets = Dataset.objects.filter(param).order_by('num')
+        return datasets

exports/scripts/generate_metadata_exports.py

@@ -1,16 +1,16 @@
 from datetime import date
 from omicspred.models import Dataset
 from exports.metadata_build_export import MetadataExport
+from exports.datasets import DatasetsSelection
 from exports.config import metadata_exports_dir, metadata_exports_publication_id, sqlite_exports_dir
 from django.db.models import Q
 
 
 def run():
-    # datasets = Dataset.objects.filter(Q(publication_id=metadata_exports_publication_id) & Q(name__icontains='sQTL - Enet')).order_by('num')
-    # datasets = Dataset.objects.filter(Q(publication_id=metadata_exports_publication_id) & Q(num__lte=56)).order_by('num')
-    datasets = Dataset.objects.filter(publication_id=metadata_exports_publication_id).order_by('num')
-    # datasets = Dataset.objects.all().order_by('num')
-    # datasets = Dataset.objects.filter(num=6).order_by('num')
+    # Fetch dataset(s)
+    ds_selection = DatasetsSelection()
+    datasets = ds_selection.get_datasets()
+
     print("## Start metadata exports, dataset by dataset")
     datasets_total = len(datasets)
     count_dataset = 0

exports/scripts/generate_phewas_exports.py

@@ -1,20 +1,16 @@
 from omicspred.models import Dataset
 from exports.phewas_export import PheWASExport
+from exports.datasets import DatasetsSelection
 from exports.config import phewas_exports_dir
-# from django.db.models import Q
 
 
 # Need to run first the script import/phewas/add_mappings (and store the SQLite file in <raw_file_dir>/metadata)
 
 def run():
-    # datasets = Dataset.objects.filter(Q(publication_id=metadata_exports_publication_id) & Q(name__icontains='sQTL - Enet')).order_by('num')
-    # datasets = Dataset.objects.filter(Q(publication_id=metadata_exports_publication_id) & Q(num__lte=56)).order_by('num')
-    # datasets = Dataset.objects.filter(publication_id=metadata_exports_publication_id).order_by('num')
-    # datasets = Dataset.objects.all().order_by('num')
-    datasets = Dataset.objects.filter(publication_id=6).order_by('num')
-    # datasets = Dataset.objects.filter(id__in=['OPD000056','OPD000105','OPD000154','OPD000203'])
-    # datasets = Dataset.objects.filter(id__in=['OPD000204','OPD000205','OPD000208','OPD000209','OPD000210','OPD000211','OPD000212','OPD000213']).order_by('num')
-    # datasets = Dataset.objects.filter(num__in=[2]).order_by('num')
+    # Fetch dataset(s)
+    ds_selection = DatasetsSelection()
+    datasets = ds_selection.get_datasets()
+
     print("## Start PheWAS exports, dataset by dataset")
     datasets_total = len(datasets)
     count_dataset = 0
@@ -26,10 +22,5 @@ def run():
 
         # Prepare data for exports - only DB
         phewas_export = PheWASExport(phewas_export_file, phewas_exports_dir, dataset)
-
-        # Prepare data for exports - DB + file
-        # raw_file_dir = '/Users/lg10/Workspace/datafiles/OmicsPred/phenotypes'
-        # phewas_exports_dir_all = f'{phewas_exports_dir}/all'
-        # phewas_export = PheWASExport(phewas_export_file, phewas_exports_dir_all, dataset, raw_file_dir)
 
         phewas_export.generate_export()

exports/scripts/generate_sqlite_files.py

@@ -1,5 +1,6 @@
 from omicspred.models import *
 from exports.sqlite_export import SqliteExport
+from exports.datasets import DatasetsSelection
 from exports.config import sqlite_default_values
 
 
@@ -98,7 +99,8 @@ def run(*args):
             skip_zip = sqlite_default_values['skip_zip']
 
     # Fetch dataset(s)
-    datasets = Dataset.objects.filter(publication__id=opp_id).order_by('id')
+    ds_selection = DatasetsSelection()
+    datasets = ds_selection.get_datasets()
     # Create SqliteExport object
     sqlite_export = SqliteExport(opp_id,output_sqlite_dir,scoring_files_dir,datasets,use_different_id_as_gene)
     # Generate SQLite export file(s)