[Fix] Stereochemistry prediction handling (#228 and #238) (#250)

**This pull request will need a separate migration pull-request**

I have added an alert box in two places when the user tries to predict with stereo chemistry.

When a user predicts a pathway with stereo chemistry an alert box is shown in that node's hover.
To do this I added two new fields. Pathway now has a "predicted" BooleanField indicating whether it was predicted or not. It is set to True if the pathway mode for prediction is "predict" or "incremental" and False if it is "build". I think it is a flag that could be useful in the future, perhaps for analysing how many predicted pathways are in enviPath?
Node now has a `stereo_removed` BooleanField which is set to True if the Node's parent Pathways has "predicted" as true and the node SMILES has stereochemistry.
<img width="500" alt="{927AC9FF-DBC9-4A19-9E6E-0EDD3B08C7AC}.png" src="attachments/69ea29bc-c2d2-4cd2-8e98-aae5c5737f69">

When a user does a prediction on a model's page it shows at the top of the list. This did not require any new fields as the entered SMILES does not get saved anywhere.
<img width="500" alt="{BED66F12-5F07-419E-AAA6-FE1FE5B4F266}.png" src="attachments/5fcc3a9b-4d1a-4e48-acac-76b7571f6507">

I think the alert box is an alright solution but if you have a great idea for something that looks/fits better please change it or let me know.

Co-authored-by: Tim Lorsbach <tim@lorsba.ch>
Reviewed-on: enviPath/enviPy#250
Co-authored-by: Liam Brydon <lbry121@aucklanduni.ac.nz>
Co-committed-by: Liam Brydon <lbry121@aucklanduni.ac.nz>

.gitea/workflows/ci.yaml

@@ -41,7 +41,7 @@ jobs:
       EP_DATA_DIR: /opt/enviPy/
       ALLOWED_HOSTS: 127.0.0.1,localhost
       DEBUG: True
-      LOG_LEVEL: DEBUG
+      LOG_LEVEL: INFO
       MODEL_BUILDING_ENABLED: True
       APPLICABILITY_DOMAIN_ENABLED: True
       ENVIFORMER_PRESENT: True

epdb/migrations/0012_node_stereo_removed_pathway_predicted.py

@@ -0,0 +1,22 @@
+# Generated by Django 5.2.7 on 2025-12-02 13:09
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+    dependencies = [
+        ("epdb", "0011_auto_20251111_1413"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="node",
+            name="stereo_removed",
+            field=models.BooleanField(default=False),
+        ),
+        migrations.AddField(
+            model_name="pathway",
+            name="predicted",
+            field=models.BooleanField(default=False),
+        ),
+    ]

epdb/models.py

@@ -1593,6 +1593,7 @@ class Pathway(EnviPathModel, AliasMixin, ScenarioMixin):
     setting = models.ForeignKey(
         "epdb.Setting", verbose_name="Setting", on_delete=models.CASCADE, null=True, blank=True
     )
+    predicted = models.BooleanField(default=False, null=False)
 
     @property
     def root_nodes(self):
@@ -1804,6 +1805,7 @@ class Pathway(EnviPathModel, AliasMixin, ScenarioMixin):
         smiles: str,
         name: Optional[str] = None,
         description: Optional[str] = None,
+        predicted: bool = False,
     ):
         pw = Pathway()
         pw.package = package
@@ -1816,6 +1818,7 @@ class Pathway(EnviPathModel, AliasMixin, ScenarioMixin):
         pw.name = name
         if description is not None and description.strip() != "":
             pw.description = nh3.clean(description, tags=s.ALLOWED_HTML_TAGS).strip()
+        pw.predicted = predicted
 
         pw.save()
         try:
@@ -1946,6 +1949,7 @@ class Node(EnviPathModel, AliasMixin, ScenarioMixin):
     )
     out_edges = models.ManyToManyField("epdb.Edge", verbose_name="Outgoing Edges")
     depth = models.IntegerField(verbose_name="Node depth", null=False, blank=False)
+    stereo_removed = models.BooleanField(default=False, null=False)
 
     def _url(self):
         return "{}/node/{}".format(self.pathway.url, self.uuid)
@@ -1955,6 +1959,7 @@ class Node(EnviPathModel, AliasMixin, ScenarioMixin):
 
         return {
             "depth": self.depth,
+            "stereo_removed": self.stereo_removed,
             "url": self.url,
             "node_label_id": self.default_node_label.url,
             "image": f"{self.url}?image=svg",
@@ -1980,12 +1985,17 @@ class Node(EnviPathModel, AliasMixin, ScenarioMixin):
         name: Optional[str] = None,
         description: Optional[str] = None,
     ):
+        stereo_removed = False
+        if pathway.predicted and FormatConverter.has_stereo(smiles):
+            smiles = FormatConverter.standardize(smiles, remove_stereo=True)
+            stereo_removed = True
         c = Compound.create(pathway.package, smiles, name=name, description=description)
 
         if Node.objects.filter(pathway=pathway, default_node_label=c.default_structure).exists():
             return Node.objects.get(pathway=pathway, default_node_label=c.default_structure)
 
         n = Node()
+        n.stereo_removed = stereo_removed
         n.pathway = pathway
         n.depth = depth
 

epdb/views.py

@@ -961,9 +961,9 @@ def package_model(request, package_uuid, model_uuid):
             # Check if smiles is non empty and valid
             if smiles == "":
                 return JsonResponse({"error": "Received empty SMILES"}, status=400)
-
+            stereo = FormatConverter.has_stereo(smiles)
             try:
-                stand_smiles = FormatConverter.standardize(smiles)
+                stand_smiles = FormatConverter.standardize(smiles, remove_stereo=True)
             except ValueError:
                 return JsonResponse({"error": f'"{smiles}" is not a valid SMILES'}, status=400)
 
@@ -974,7 +974,7 @@ def package_model(request, package_uuid, model_uuid):
                     current_user, predict_simple, current_model.pk, stand_smiles
                 )
 
-                res = []
+                res = {"pred": [], "stereo": stereo}
 
                 for pr in pred_res:
                     if len(pr) > 0:
@@ -983,7 +983,7 @@ def package_model(request, package_uuid, model_uuid):
                             logger.debug(f"Checking {prod_set}")
                             products.append(tuple([x for x in prod_set]))
 
-                        res.append(
+                        res["pred"].append(
                             {
                                 "products": list(set(products)),
                                 "probability": pr.probability,
@@ -1931,7 +1931,6 @@ def package_pathways(request, package_uuid):
                 "Pathway prediction failed!",
                 "Pathway prediction failed due to missing or empty SMILES",
             )
-
         try:
             stand_smiles = FormatConverter.standardize(smiles)
         except ValueError:
@@ -1954,8 +1953,13 @@ def package_pathways(request, package_uuid):
             prediction_setting = SettingManager.get_setting_by_url(current_user, prediction_setting)
         else:
             prediction_setting = current_user.prediction_settings()
-
+        pw = Pathway.create(
-        pw = Pathway.create(current_package, stand_smiles, name=name, description=description)
+            current_package,
+            stand_smiles,
+            name=name,
+            description=description,
+            predicted=pw_mode in {"predict", "incremental"},
+        )
 
         # set mode
         pw.kv.update({"mode": pw_mode})

static/js/pw.js

@@ -360,7 +360,11 @@ function draw(pathway, elem) {
     }
 
     function node_popup(n) {
-        popupContent = "<a href='" + n.url + "'>" + n.name + "</a><br>";
+        popupContent = "";
+        if (n.stereo_removed) {
+            popupContent += "<span class='alert alert-warning alert-soft'>Removed stereochemistry for prediction</span>";
+        }
+        popupContent += "<a href='" + n.url + "'>" + n.name + "</a><br>";
         popupContent += "Depth " + n.depth + "<br>"
 
         if (appDomainViewEnabled) {

templates/objects/model.html

@@ -117,7 +117,9 @@
       <!-- Predict Panel -->
       <div class="collapse-arrow bg-base-200 collapse">
         <input type="checkbox" checked />
-        <div class="collapse-title text-xl font-medium">Predict</div>
+        <div class="collapse-title text-xl font-medium" id="predictTitle">
+          Predict
+        </div>
         <div class="collapse-content">
           <div class="form-control">
             <div class="join w-full">
@@ -193,7 +195,13 @@
     {# FIXME: This is a hack to get the precision recall curve data into the JavaScript code. #}
     <script>
         function handlePredictionResponse(data) {
-            let res = "<table class='table table-zebra'>"
+            let stereo = data["stereo"]
+            data = data["pred"]
+            let res = ""
+            if (stereo) {
+                res += "<span class='alert alert-warning alert-soft'>Removed stereochemistry for prediction</span><br>"
+            }
+            res += "<table class='table table-zebra'>"
             res += "<thead>"
             res += "<th scope='col'>#</th>"
 

tests/test_compound_model.py

@@ -1,11 +1,12 @@
-from django.test import TestCase
+from django.conf import settings as s
-
+from django.test import TestCase, override_settings
 from epdb.logic import PackageManager
 from epdb.models import Compound, User, CompoundStructure
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class CompoundTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     def setUp(self):
         pass

tests/test_copy_objects.py

@@ -1,12 +1,14 @@
-from django.test import TestCase
+from django.conf import settings as s
+from django.test import TestCase, override_settings
 
 
 from epdb.logic import PackageManager
 from epdb.models import Compound, User, Reaction
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class CopyTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/test_dataset.py

@@ -1,14 +1,16 @@
 import os.path
 from tempfile import TemporaryDirectory
-from django.test import TestCase
+from django.conf import settings as s
+from django.test import TestCase, override_settings
 from epdb.logic import PackageManager
 from epdb.models import Reaction, Compound, User, Rule, Package
 from utilities.chem import FormatConverter
 from utilities.ml import RuleBasedDataset, EnviFormerDataset
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class DatasetTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     def setUp(self):
         self.cs1 = Compound.create(
@@ -106,7 +108,9 @@ class DatasetTest(TestCase):
     def test_extra_features(self):
         reactions = [r for r in Reaction.objects.filter(package=self.BBD_SUBSET)]
         applicable_rules = [r for r in Rule.objects.filter(package=self.BBD_SUBSET)]
-        ds = RuleBasedDataset.generate_dataset(reactions, applicable_rules, feat_funcs=[FormatConverter.maccs, FormatConverter.morgan])
+        ds = RuleBasedDataset.generate_dataset(
+            reactions, applicable_rules, feat_funcs=[FormatConverter.maccs, FormatConverter.morgan]
+        )
         print(ds.shape)
 
     def test_to_arff(self):

tests/test_enviformer.py

@@ -3,7 +3,8 @@ from datetime import datetime
 from tempfile import TemporaryDirectory
 
 from django.conf import settings as s
-from django.test import TestCase, tag
+from django.test import TestCase, override_settings, tag
+
 
 from epdb.logic import PackageManager
 from epdb.models import EnviFormer, Setting, User
@@ -30,8 +31,9 @@ def measure_predict(mod, pathway_pk=None):
 
 
 @tag("slow")
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class EnviFormerTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/test_model.py

@@ -2,7 +2,7 @@ from tempfile import TemporaryDirectory
 
 import numpy as np
 from django.conf import settings as s
-from django.test import TestCase
+from django.test import TestCase, override_settings
 
 from epdb.logic import PackageManager
 from epdb.models import MLRelativeReasoning, RuleBasedRelativeReasoning, User
@@ -10,8 +10,9 @@ from epdb.models import MLRelativeReasoning, RuleBasedRelativeReasoning, User
 Package = s.GET_PACKAGE_MODEL()
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class ModelTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):
@@ -52,7 +53,7 @@ class ModelTest(TestCase):
                 expected = {
                     ("CC=O", "CCNC(=O)C1=CC(C)=CC=C1"): (
                         "bt0243-4301",
-                        np.float64(0.33333333333333337),
+                        np.float64(0.5),
                     ),
                     ("CC1=CC=CC(C(=O)O)=C1", "CCNCC"): ("bt0430-4011", np.float64(0.25)),
                 }

tests/test_multigen_eval.py

@@ -1,5 +1,5 @@
 from django.conf import settings as s
-from django.test import TestCase
+from django.test import TestCase, override_settings
 from networkx.utils.misc import graphs_equal
 
 from epdb.logic import PackageManager, SPathway
@@ -9,8 +9,9 @@ from utilities.ml import graph_from_pathway, multigen_eval, pathway_edit_eval
 Package = s.GET_PACKAGE_MODEL()
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class MultiGenTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/test_reaction_model.py

@@ -1,11 +1,13 @@
-from django.test import TestCase
+from django.conf import settings as s
+from django.test import TestCase, override_settings
 
 from epdb.logic import PackageManager
 from epdb.models import Compound, User, Reaction, Rule
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class ReactionTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/test_rule_model.py

@@ -1,11 +1,12 @@
-from django.test import TestCase
+from django.conf import settings as s
-
+from django.test import TestCase, override_settings
 from epdb.logic import PackageManager
 from epdb.models import Rule, User
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class RuleTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/test_simpleambitrule.py

@@ -1,15 +1,16 @@
 from unittest.mock import MagicMock, PropertyMock, patch
 
 from django.conf import settings as s
-from django.test import TestCase
+from django.test import TestCase, override_settings
 
 from epdb.logic import PackageManager
 from epdb.models import SimpleAmbitRule, User
 from utilities.chem import FormatConverter
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class SimpleAmbitRuleTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/views/test_compound_views.py

@@ -1,4 +1,5 @@
-from django.test import TestCase
+from django.conf import settings as s
+from django.test import TestCase, override_settings
 from django.urls import reverse
 from envipy_additional_information import Temperature, Interval
 
@@ -6,8 +7,9 @@ from epdb.logic import UserManager, PackageManager
 from epdb.models import Compound, Scenario, ExternalDatabase
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class CompoundViewTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/views/test_model_views.py

@@ -9,12 +9,12 @@ Package = s.GET_PACKAGE_MODEL()
 
 
 @override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
-class PathwayViewTest(TestCase):
+class ModelViewTest(TestCase):
     fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):
-        super(PathwayViewTest, cls).setUpClass()
+        super(ModelViewTest, cls).setUpClass()
         cls.user1 = UserManager.create_user(
             "user1",
             "user1@envipath.com",
@@ -72,7 +72,7 @@ class PathwayViewTest(TestCase):
             },
         ]
 
-        actual = response.json()
+        actual = response.json()["pred"]
         self.assertEqual(actual, expected)
 
         response = self.client.get(

tests/views/test_package_views.py

@@ -1,6 +1,6 @@
 from django.conf import settings as s
 from django.core.files.uploadedfile import SimpleUploadedFile
-from django.test import TestCase, tag
+from django.test import TestCase, override_settings, tag
 from django.urls import reverse
 
 from epdb.logic import UserManager
@@ -15,8 +15,9 @@ from epdb.models import (
 Package = s.GET_PACKAGE_MODEL()
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class PackageViewTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/views/test_reaction_views.py

@@ -1,4 +1,5 @@
-from django.test import TestCase
+from django.conf import settings as s
+from django.test import TestCase, override_settings
 from django.urls import reverse
 from envipy_additional_information import Temperature, Interval
 
@@ -6,8 +7,9 @@ from epdb.logic import UserManager, PackageManager
 from epdb.models import Reaction, Scenario, ExternalDatabase
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class ReactionViewTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/views/test_rule_views.py

@@ -1,4 +1,5 @@
-from django.test import TestCase
+from django.conf import settings as s
+from django.test import TestCase, override_settings
 from django.urls import reverse
 from envipy_additional_information import Temperature, Interval
 
@@ -6,8 +7,9 @@ from epdb.logic import UserManager, PackageManager
 from epdb.models import Rule, Scenario
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class RuleViewTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):

tests/views/test_user_views.py

@@ -1,12 +1,14 @@
-from django.test import TestCase
+from django.conf import settings as s
+from django.test import TestCase, override_settings
+from django.urls import reverse
 
 from epdb.logic import PackageManager
 from epdb.models import Package, User
-from django.urls import reverse
 
 
+@override_settings(MODEL_DIR=s.FIXTURE_DIRS[0] / "models", CELERY_TASK_ALWAYS_EAGER=True)
 class UserViewTest(TestCase):
-    fixtures = ["test_fixtures.jsonl.gz"]
+    fixtures = ["test_fixtures_incl_model.jsonl.gz"]
 
     @classmethod
     def setUpClass(cls):
@@ -48,8 +50,9 @@ class UserViewTest(TestCase):
             },
         )
         self.assertEqual(response.status_code, 200)
-        # TODO currently fails as the fixture does not provide a global setting...
+        self.assertContains(
-        self.assertContains(response, "Registration failed!")
+            response, "Your account has been created! An admin will activate it soon!"
+        )
 
     def test_register_password_mismatch(self):
         response = self.client.post(

utilities/chem.py

@@ -81,6 +81,10 @@ class FormatConverter(object):
     def formula(smiles):
         return Chem.rdMolDescriptors.CalcMolFormula(FormatConverter.from_smiles(smiles))
 
+    @staticmethod
+    def has_stereo(smiles):
+        return "@" in smiles or "/" in smiles or "\\" in smiles
+
     @staticmethod
     def from_smiles(smiles):
         return Chem.MolFromSmiles(smiles)