From 4be97eb540f7a887a7effda3044853fcf669c620 Mon Sep 17 00:00:00 2001
From: Helle van den Maagdenberg <h.w.van.den.maagdenberg@lacdr.leidenuniv.nl>
Date: Sun, 10 Mar 2024 17:50:09 +0100
Subject: [PATCH 1/5] fix bug with features unordered and standardizer refit
 for applicability domain

---
 qsprpred/models/models.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/qsprpred/models/models.py b/qsprpred/models/models.py
index 169bbd1f..85d78251 100644
--- a/qsprpred/models/models.py
+++ b/qsprpred/models/models.py
@@ -562,7 +562,7 @@ def predictMols(
         # return predictions and if mols are within applicability domain if requested
         if hasattr(self, "applicabilityDomain") and use_applicability_domain:
             in_domain = self.applicabilityDomain.contains(
-                dataset.getFeatures(concat=True)
+                dataset.getFeatures(concat=True, ordered=True, refit_standardizer=False)
             )
             in_domain = self.handleInvalidsInPredictions(mols, in_domain, failed_mask)
             return predictions, in_domain.values

From 4b1ea333e7d4be3c4dde780f8409671bdad0cd08 Mon Sep 17 00:00:00 2001
From: Helle van den Maagdenberg <h.w.van.den.maagdenberg@lacdr.leidenuniv.nl>
Date: Mon, 11 Mar 2024 13:28:14 +0100
Subject: [PATCH 2/5] add unit test for preds with ap

---
 qsprpred/data/processing/tests.py |   4 +-
 qsprpred/models/tests.py          | 152 +++++++++++++++++++++---------
 2 files changed, 111 insertions(+), 45 deletions(-)

diff --git a/qsprpred/data/processing/tests.py b/qsprpred/data/processing/tests.py
index 536c3898..078cd626 100644
--- a/qsprpred/data/processing/tests.py
+++ b/qsprpred/data/processing/tests.py
@@ -287,9 +287,7 @@ def setUp(self):
     def testApplicabilityDomain(self):
         """Test the applicability domain fitting, transforming and serialization."""
         ad = MLChemADWrapper(
-            KNNApplicabilityDomain(
-                dist="rogerstanimoto", scaling=None, hard_threshold=0.75
-            )
+            KNNApplicabilityDomain(dist="jaccard", scaling=None, alpha=0.95)
         )
         ad.fit(self.dataset.X)
         self.assertIsInstance(ad.contains(self.dataset.X), pd.DataFrame)
diff --git a/qsprpred/models/tests.py b/qsprpred/models/tests.py
index 245aea63..b0053c42 100644
--- a/qsprpred/models/tests.py
+++ b/qsprpred/models/tests.py
@@ -5,6 +5,7 @@
 from unittest import TestCase
 
 import numpy as np
+from mlchemad.applicability_domains import KNNApplicabilityDomain
 from parameterized import parameterized
 from sklearn.cross_decomposition import PLSRegression
 from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor
@@ -13,35 +14,31 @@
     accuracy_score,
     explained_variance_score,
     log_loss,
-    top_k_accuracy_score,
+    make_scorer,
     mean_squared_error,
     roc_auc_score,
+    top_k_accuracy_score,
 )
-from sklearn.metrics import make_scorer
 from sklearn.naive_bayes import GaussianNB
 from sklearn.neighbors import KNeighborsClassifier, KNeighborsRegressor
 from sklearn.svm import SVC, SVR
 from xgboost import XGBClassifier, XGBRegressor
 
-from .assessment.classification import create_metrics_summary
-from .assessment.regression import create_correlation_summary
+from ..data.processing.applicability_domain import MLChemADWrapper
 from ..models.early_stopping import EarlyStopping, EarlyStoppingMode, early_stopping
 from ..models.metrics import SklearnMetrics
-from ..models.monitors import (
-    BaseMonitor,
-    FileMonitor,
-    ListMonitor,
-)
+from ..models.monitors import BaseMonitor, FileMonitor, ListMonitor
 from ..models.scikit_learn import SklearnModel
 from ..tasks import TargetTasks
 from ..utils.testing.base import QSPRTestCase
 from ..utils.testing.check_mixins import ModelCheckMixIn, MonitorsCheckMixIn
 from ..utils.testing.path_mixins import ModelDataSetsPathMixIn
+from .assessment.classification import create_metrics_summary
+from .assessment.regression import create_correlation_summary
 
 
 class SklearnBaseModelTestCase(ModelDataSetsPathMixIn, ModelCheckMixIn, QSPRTestCase):
     """This class holds the tests for the SklearnModel class."""
-
     def setUp(self):
         super().setUp()
         self.setUpPaths()
@@ -78,17 +75,14 @@ def getModel(
 
 class TestSklearnRegression(SklearnBaseModelTestCase):
     """Test the SklearnModel class for regression models."""
-
     @parameterized.expand(
         [
             (alg_name, TargetTasks.REGRESSION, alg_name, alg, random_state)
             for alg, alg_name in (
                 (RandomForestRegressor, "RFR"),
                 (XGBRegressor, "XGBR"),
-            )
-            for random_state in ([None], [1, 42], [42, 42])
-        ]
-        + [
+            ) for random_state in ([None], [1, 42], [42, 42])
+        ] + [
             (alg_name, TargetTasks.REGRESSION, alg_name, alg, [None])
             for alg, alg_name in (
                 (PLSRegression, "PLSR"),
@@ -105,7 +99,10 @@ def testRegressionBasicFit(self, _, task, model_name, model_class, random_state)
             parameters = None
         # initialize dataset
         dataset = self.createLargeTestDataSet(
-            target_props=[{"name": "CL", "task": task}],
+            target_props=[{
+                "name": "CL",
+                "task": task
+            }],
             preparation_settings=self.getDefaultPrep(),
         )
         # initialize model for training from class
@@ -149,7 +146,10 @@ def testPLSRegressionSummaryWithSeed(self):
         model_class = PLSRegression
         parameters = None
         dataset = self.createLargeTestDataSet(
-            target_props=[{"name": "CL", "task": task}],
+            target_props=[{
+                "name": "CL",
+                "task": task
+            }],
             preparation_settings=self.getDefaultPrep(),
         )
         model = self.getModel(
@@ -177,16 +177,14 @@ def testPLSRegressionSummaryWithSeed(self):
 
 class TestSklearnRegressionMultiTask(SklearnBaseModelTestCase):
     """Test the SklearnModel class for multi-task regression models."""
-
     @parameterized.expand(
         [
             (alg_name, alg_name, alg, random_state)
-            for alg, alg_name in ((RandomForestRegressor, "RFR"),)
+            for alg, alg_name in ((RandomForestRegressor, "RFR"), )
             for random_state in ([None], [1, 42], [42, 42])
-        ]
-        + [
+        ] + [
             (alg_name, alg_name, alg, [None])
-            for alg, alg_name in ((KNeighborsRegressor, "KNNR"),)
+            for alg, alg_name in ((KNeighborsRegressor, "KNNR"), )
         ]
     )
     def testRegressionMultiTaskFit(self, _, model_name, model_class, random_state):
@@ -246,13 +244,20 @@ def testRegressionMultiTaskFit(self, _, model_name, model_class, random_state):
 class TestSklearnSerialization(SklearnBaseModelTestCase):
     def testJSON(self):
         dataset = self.createLargeTestDataSet(
-            target_props=[{"name": "CL", "task": TargetTasks.SINGLECLASS, "th": [6.5]}],
+            target_props=[{
+                "name": "CL",
+                "task": TargetTasks.SINGLECLASS,
+                "th": [6.5]
+            }],
             preparation_settings=self.getDefaultPrep(),
         )
         model = self.getModel(
             name="TestSerialization",
             alg=RandomForestClassifier,
-            parameters={"n_jobs": self.nCPU, "n_estimators": 10},
+            parameters={
+                "n_jobs": self.nCPU,
+                "n_estimators": 10
+            },
             random_state=42,
         )
         model.save()
@@ -271,28 +276,23 @@ def testJSON(self):
 
 class TestSklearnClassification(SklearnBaseModelTestCase):
     """Test the SklearnModel class for classification models."""
-
     @parameterized.expand(
         [
             (f"{alg_name}_{task}", task, th, alg_name, alg, random_state)
             for alg, alg_name in (
                 (RandomForestClassifier, "RFC"),
                 (XGBClassifier, "XGBC"),
-            )
-            for task, th in (
+            ) for task, th in (
                 (TargetTasks.SINGLECLASS, [6.5]),
                 (TargetTasks.MULTICLASS, [0, 2, 10, 1100]),
-            )
-            for random_state in ([None], [1, 42], [42, 42])
-        ]
-        + [
+            ) for random_state in ([None], [1, 42], [42, 42])
+        ] + [
             (f"{alg_name}_{task}", task, th, alg_name, alg, [None])
             for alg, alg_name in (
                 (SVC, "SVC"),
                 (KNeighborsClassifier, "KNNC"),
                 (GaussianNB, "NB"),
-            )
-            for task, th in (
+            ) for task, th in (
                 (TargetTasks.SINGLECLASS, [6.5]),
                 (TargetTasks.MULTICLASS, [0, 2, 10, 1100]),
             )
@@ -317,7 +317,11 @@ def testClassificationBasicFit(
             parameters = {"subsample": 0.3}
         # initialize dataset
         dataset = self.createLargeTestDataSet(
-            target_props=[{"name": "CL", "task": task, "th": th}],
+            target_props=[{
+                "name": "CL",
+                "task": task,
+                "th": th
+            }],
             preparation_settings=self.getDefaultPrep(),
         )
         # test classifier
@@ -361,7 +365,11 @@ def testRandomForestClassifierFitWithSeed(self):
         }
         # initialize dataset
         dataset = self.createLargeTestDataSet(
-            target_props=[{"name": "CL", "task": TargetTasks.SINGLECLASS, "th": [6.5]}],
+            target_props=[{
+                "name": "CL",
+                "task": TargetTasks.SINGLECLASS,
+                "th": [6.5]
+            }],
             preparation_settings=self.getDefaultPrep(),
         )
         # test classifier
@@ -391,16 +399,14 @@ def testRandomForestClassifierFitWithSeed(self):
 
 class TestSklearnClassificationMultiTask(SklearnBaseModelTestCase):
     """Test the SklearnModel class for multi-task classification models."""
-
     @parameterized.expand(
         [
             (alg_name, alg_name, alg, random_state)
-            for alg, alg_name in ((RandomForestClassifier, "RFC"),)
+            for alg, alg_name in ((RandomForestClassifier, "RFC"), )
             for random_state in ([None], [1, 42], [42, 42])
-        ]
-        + [
+        ] + [
             (alg_name, alg_name, alg, [None])
-            for alg, alg_name in ((KNeighborsClassifier, "KNNC"),)
+            for alg, alg_name in ((KNeighborsClassifier, "KNNC"), )
         ]
     )
     def testClassificationMultiTaskFit(self, _, model_name, model_class, random_state):
@@ -470,7 +476,6 @@ def testClassificationMultiTaskFit(self, _, model_name, model_class, random_stat
 
 class TestMetrics(TestCase):
     """Test the SklearnMetrics from the metrics module."""
-
     def test_SklearnMetrics(self):
         """Test the sklearn metrics wrapper."""
 
@@ -648,7 +653,6 @@ def test_EarlyStopping(self):
 
     def test_early_stopping_decorator(self):
         """Test the early stopping decorator."""
-
         class test_class:
             def __init__(self, support=True):
                 self.earlyStopping = EarlyStopping(EarlyStoppingMode.RECORDING)
@@ -752,3 +756,67 @@ def testListMonitor(self):
             False,
             [BaseMonitor(), FileMonitor()],
         )
+
+
+class TestAttachedApplicabilityDomain(ModelDataSetsPathMixIn, QSPRTestCase):
+    def setUp(self):
+        super().setUp()
+        self.setUpPaths()
+
+    def testAttachedApplicabilityDomain(self):
+        """Test the attached applicability domain class."""
+
+        # initialize test dataset with attached applicability domain
+        dataset = self.createLargeTestDataSet(
+            target_props=[{
+                "name": "CL",
+                "task": "REGRESSION"
+            }],
+            preparation_settings={
+                **self.getDefaultPrep(),
+                "applicability_domain":
+                    KNNApplicabilityDomain(dist="euclidean", alpha=0.9, scaling=None),
+            },
+        )
+        # initialize model for training
+        model = SklearnModel(
+            base_dir=self.generatedModelsPath,
+            alg=RandomForestRegressor,
+            name="RFR_with_AD",
+            parameters={"n_jobs": self.nCPU},
+            random_state=42,
+        )
+
+        model.fitDataset(dataset)
+
+        # check if the applicability domain is attached to the model
+        self.assertTrue(hasattr(model, "applicabilityDomain"))
+        self.assertIsInstance(model.applicabilityDomain, MLChemADWrapper)
+
+        # check if the applicability domain is saved and loaded correctly
+        model.save()
+        model2 = SklearnModel.fromFile(model.metaFile)
+        self.assertTrue(hasattr(model2, "applicabilityDomain"))
+        self.assertIsInstance(model2.applicabilityDomain, MLChemADWrapper)
+
+        # make predictions with mlchemad ap on the dataset directly
+        comparison_ap = KNNApplicabilityDomain(
+            dist="euclidean", alpha=0.9, scaling=None
+        )
+        features = dataset.getFeatures(
+            concat=True, ordered=True, refit_standardizer=False
+        )
+        comparison_ap.fit(features)
+        ap_pred = comparison_ap.contains(features)
+
+        # check if the applicability domain predictions from the dataset are equal to the ones from the model
+        _, ap_preds_model = model.predictMols(
+            dataset.df["SMILES"], use_applicability_domain=True
+        )
+        self.assertTrue(np.array_equal(ap_pred.reshape(-1, 1), ap_preds_model))
+
+        # check if the applicability domain predictions arrays are equal after saving and loading
+        _, ap_preds_model2 = model2.predictMols(
+            dataset.df["SMILES"], use_applicability_domain=True
+        )
+        self.assertTrue(np.array_equal(ap_pred.reshape(-1, 1), ap_preds_model2))

From 8113fd2aac87b3ff9c2763f4b896fe34bcf610e2 Mon Sep 17 00:00:00 2001
From: Helle van den Maagdenberg <h.w.van.den.maagdenberg@lacdr.leidenuniv.nl>
Date: Mon, 11 Mar 2024 13:30:31 +0100
Subject: [PATCH 3/5] update changelog

---
 CHANGELOG.md | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 7d5ee55f..07ea604c 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,15 +1,15 @@
 # Change Log
 
-From v3.0.1 to v3.0.2
+From v3.0.2 to v3.0.1
 
 ## Fixes
 
-- Fixed a bug with incorrect checking of whether a data set contains descriptors or not.
+- Fixed a bug where an attached standardizer would be refit when calling 
+`QSPRModel.predictMols` with `use_applicability_domain=True`.
 
 ## Changes
 
-- If a `MoleculeTable` contains descriptors and `QSPRDataset.fromMolTable` is called the
-  descriptors are now automatically loaded as features of the created data set.
+None.
 
 ## New Features
 

From 2150d00de9038a6875b988494d31c7f87ae10c74 Mon Sep 17 00:00:00 2001
From: Helle van den Maagdenberg <h.w.van.den.maagdenberg@lacdr.leidenuniv.nl>
Date: Mon, 11 Mar 2024 13:30:51 +0100
Subject: [PATCH 4/5] fix typo in changelog

---
 CHANGELOG.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 07ea604c..d1d5798a 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,6 +1,6 @@
 # Change Log
 
-From v3.0.2 to v3.0.1
+From v3.0.2 to v3.0.3
 
 ## Fixes
 

From 224f0359a937ae60f7b6c40cf8918a42cb3647d3 Mon Sep 17 00:00:00 2001
From: Helle van den Maagdenberg <h.w.van.den.maagdenberg@lacdr.leidenuniv.nl>
Date: Mon, 11 Mar 2024 14:07:50 +0100
Subject: [PATCH 5/5] rerun ap tutorial

---
 .../basics/data/applicability_domain.ipynb    | 247 ++++++++----------
 1 file changed, 103 insertions(+), 144 deletions(-)

diff --git a/tutorials/basics/data/applicability_domain.ipynb b/tutorials/basics/data/applicability_domain.ipynb
index bde885c4..3d9c8e5a 100644
--- a/tutorials/basics/data/applicability_domain.ipynb
+++ b/tutorials/basics/data/applicability_domain.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -23,18 +23,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Failed to find the pandas get_adjustment() function to patch\n",
-      "Failed to patch pandas - PandasTools will have limited functionality\n",
-      "qsprpred - WARNING - Existing data set found, but also found a data frame in store. Refusing to overwrite data. If you want to overwrite data in store, set overwrite=True.\n"
-     ]
-    },
     {
      "data": {
       "text/html": [
@@ -61,8 +52,6 @@
        "      <th>Year</th>\n",
        "      <th>QSPRID</th>\n",
        "      <th>pchembl_value_Mean_original</th>\n",
-       "      <th>Split_IsOutlier</th>\n",
-       "      <th>Split_IsTrain</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>QSPRID</th>\n",
@@ -71,30 +60,24 @@
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
-       "      <th></th>\n",
-       "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_0000</th>\n",
-       "      <td>Cc1cc(C)n(-c2cc(NC(=O)CCN(C)C)nc(-c3ccc(C)o3)n...</td>\n",
+       "      <td>Cc1nn(-c2cc(NC(=O)CCN(C)C)nc(-c3ccc(C)o3)n2)c(...</td>\n",
        "      <td>8.68</td>\n",
        "      <td>2008.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_0000</td>\n",
        "      <td>8.68</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_0001</th>\n",
-       "      <td>Nc1c(C(=O)Nc2ccc([N+](=O)[O-])cc2)sc2nc3c(cc12...</td>\n",
+       "      <td>Nc1c(C(=O)Nc2ccc([N+](=O)[O-])cc2)sc2c1cc1CCCC...</td>\n",
        "      <td>4.82</td>\n",
        "      <td>2010.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_0001</td>\n",
        "      <td>4.82</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_0002</th>\n",
@@ -103,18 +86,14 @@
        "      <td>2009.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_0002</td>\n",
        "      <td>5.65</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_0003</th>\n",
-       "      <td>CNC(=O)C12CC1C(n1cnc3c(NCc4cccc(Cl)c4)nc(C#CCC...</td>\n",
+       "      <td>CNC(=O)C12CC1C(n1cnc3c1nc(C#CCCCCC(=O)OC)nc3NC...</td>\n",
        "      <td>5.45</td>\n",
        "      <td>2009.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_0003</td>\n",
        "      <td>5.45</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_0004</th>\n",
@@ -123,8 +102,6 @@
        "      <td>2019.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_0004</td>\n",
        "      <td>5.20</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>...</th>\n",
@@ -133,8 +110,6 @@
        "      <td>...</td>\n",
        "      <td>...</td>\n",
        "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_4077</th>\n",
@@ -143,8 +118,6 @@
        "      <td>2018.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_4077</td>\n",
        "      <td>7.09</td>\n",
-       "      <td>False</td>\n",
-       "      <td>False</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_4078</th>\n",
@@ -153,18 +126,14 @@
        "      <td>2008.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_4078</td>\n",
        "      <td>8.22</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_4079</th>\n",
-       "      <td>Nc1nc(CSc2nnc(N)s2)nc(Nc2ccc(F)cc2)n1</td>\n",
+       "      <td>Nc1nc(Nc2ccc(F)cc2)nc(CSc2nnc(N)s2)n1</td>\n",
        "      <td>4.89</td>\n",
        "      <td>2010.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_4079</td>\n",
        "      <td>4.89</td>\n",
-       "      <td>True</td>\n",
-       "      <td>False</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_4080</th>\n",
@@ -173,8 +142,6 @@
        "      <td>2013.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_4080</td>\n",
        "      <td>6.51</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_4081</th>\n",
@@ -183,26 +150,24 @@
        "      <td>2014.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_4081</td>\n",
        "      <td>7.35</td>\n",
-       "      <td>False</td>\n",
-       "      <td>False</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "<p>4082 rows × 7 columns</p>\n",
+       "<p>4082 rows × 5 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
        "                                                                                    SMILES  \\\n",
        "QSPRID                                                                                       \n",
-       "ApplicabilityDomainTutorialDataset_0000  Cc1cc(C)n(-c2cc(NC(=O)CCN(C)C)nc(-c3ccc(C)o3)n...   \n",
-       "ApplicabilityDomainTutorialDataset_0001  Nc1c(C(=O)Nc2ccc([N+](=O)[O-])cc2)sc2nc3c(cc12...   \n",
+       "ApplicabilityDomainTutorialDataset_0000  Cc1nn(-c2cc(NC(=O)CCN(C)C)nc(-c3ccc(C)o3)n2)c(...   \n",
+       "ApplicabilityDomainTutorialDataset_0001  Nc1c(C(=O)Nc2ccc([N+](=O)[O-])cc2)sc2c1cc1CCCC...   \n",
        "ApplicabilityDomainTutorialDataset_0002   O=C(Nc1nc2ncccc2n2c(=O)n(-c3ccccc3)nc12)c1ccccc1   \n",
-       "ApplicabilityDomainTutorialDataset_0003  CNC(=O)C12CC1C(n1cnc3c(NCc4cccc(Cl)c4)nc(C#CCC...   \n",
+       "ApplicabilityDomainTutorialDataset_0003  CNC(=O)C12CC1C(n1cnc3c1nc(C#CCCCCC(=O)OC)nc3NC...   \n",
        "ApplicabilityDomainTutorialDataset_0004  CCCn1c(=O)c2c(nc3cc(OC)ccn32)n(CCCNC(=O)c2ccc(...   \n",
        "...                                                                                    ...   \n",
        "ApplicabilityDomainTutorialDataset_4077     CNc1ncc(C(=O)NCc2ccc(OC)cc2)c2nc(-c3ccco3)nn12   \n",
        "ApplicabilityDomainTutorialDataset_4078         Nc1nc(-c2ccco2)c2ncn(C(=O)NCCc3ccccc3)c2n1   \n",
-       "ApplicabilityDomainTutorialDataset_4079              Nc1nc(CSc2nnc(N)s2)nc(Nc2ccc(F)cc2)n1   \n",
+       "ApplicabilityDomainTutorialDataset_4079              Nc1nc(Nc2ccc(F)cc2)nc(CSc2nnc(N)s2)n1   \n",
        "ApplicabilityDomainTutorialDataset_4080      CCCOc1ccc(C=Cc2cc3c(c(=O)n(C)c(=O)n3C)n2C)cc1   \n",
        "ApplicabilityDomainTutorialDataset_4081           CCOC(=O)c1cnc(NCC(C)C)n2nc(-c3ccco3)nc12   \n",
        "\n",
@@ -234,38 +199,24 @@
        "ApplicabilityDomainTutorialDataset_4080  ApplicabilityDomainTutorialDataset_4080   \n",
        "ApplicabilityDomainTutorialDataset_4081  ApplicabilityDomainTutorialDataset_4081   \n",
        "\n",
-       "                                         pchembl_value_Mean_original  \\\n",
-       "QSPRID                                                                 \n",
-       "ApplicabilityDomainTutorialDataset_0000                         8.68   \n",
-       "ApplicabilityDomainTutorialDataset_0001                         4.82   \n",
-       "ApplicabilityDomainTutorialDataset_0002                         5.65   \n",
-       "ApplicabilityDomainTutorialDataset_0003                         5.45   \n",
-       "ApplicabilityDomainTutorialDataset_0004                         5.20   \n",
-       "...                                                              ...   \n",
-       "ApplicabilityDomainTutorialDataset_4077                         7.09   \n",
-       "ApplicabilityDomainTutorialDataset_4078                         8.22   \n",
-       "ApplicabilityDomainTutorialDataset_4079                         4.89   \n",
-       "ApplicabilityDomainTutorialDataset_4080                         6.51   \n",
-       "ApplicabilityDomainTutorialDataset_4081                         7.35   \n",
-       "\n",
-       "                                         Split_IsOutlier  Split_IsTrain  \n",
-       "QSPRID                                                                   \n",
-       "ApplicabilityDomainTutorialDataset_0000            False           True  \n",
-       "ApplicabilityDomainTutorialDataset_0001            False           True  \n",
-       "ApplicabilityDomainTutorialDataset_0002            False           True  \n",
-       "ApplicabilityDomainTutorialDataset_0003            False           True  \n",
-       "ApplicabilityDomainTutorialDataset_0004            False           True  \n",
-       "...                                                  ...            ...  \n",
-       "ApplicabilityDomainTutorialDataset_4077            False          False  \n",
-       "ApplicabilityDomainTutorialDataset_4078            False           True  \n",
-       "ApplicabilityDomainTutorialDataset_4079             True          False  \n",
-       "ApplicabilityDomainTutorialDataset_4080            False           True  \n",
-       "ApplicabilityDomainTutorialDataset_4081            False          False  \n",
+       "                                         pchembl_value_Mean_original  \n",
+       "QSPRID                                                                \n",
+       "ApplicabilityDomainTutorialDataset_0000                         8.68  \n",
+       "ApplicabilityDomainTutorialDataset_0001                         4.82  \n",
+       "ApplicabilityDomainTutorialDataset_0002                         5.65  \n",
+       "ApplicabilityDomainTutorialDataset_0003                         5.45  \n",
+       "ApplicabilityDomainTutorialDataset_0004                         5.20  \n",
+       "...                                                              ...  \n",
+       "ApplicabilityDomainTutorialDataset_4077                         7.09  \n",
+       "ApplicabilityDomainTutorialDataset_4078                         8.22  \n",
+       "ApplicabilityDomainTutorialDataset_4079                         4.89  \n",
+       "ApplicabilityDomainTutorialDataset_4080                         6.51  \n",
+       "ApplicabilityDomainTutorialDataset_4081                         7.35  \n",
        "\n",
-       "[4082 rows x 7 columns]"
+       "[4082 rows x 5 columns]"
       ]
      },
-     "execution_count": 3,
+     "execution_count": 2,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -306,14 +257,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Number of removed compounds: 114\n"
+      "Number of removed compounds: 274\n"
      ]
     },
     {
@@ -342,8 +293,8 @@
        "      <th>Year</th>\n",
        "      <th>QSPRID</th>\n",
        "      <th>pchembl_value_Mean_original</th>\n",
-       "      <th>Split_IsTrain</th>\n",
        "      <th>Split_IsOutlier</th>\n",
+       "      <th>Split_IsTrain</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>QSPRID</th>\n",
@@ -358,14 +309,44 @@
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
+       "      <th>ApplicabilityDomainTutorialDataset_0006</th>\n",
+       "      <td>Nc1nc(-c2ccccc2)cn2cc(-c3ccco3)nc12</td>\n",
+       "      <td>6.28</td>\n",
+       "      <td>2017.0</td>\n",
+       "      <td>ApplicabilityDomainTutorialDataset_0006</td>\n",
+       "      <td>6.28</td>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>ApplicabilityDomainTutorialDataset_0017</th>\n",
+       "      <td>Cn1c(=O)c2nc(-c3ccc(OCCCN4CCN(Cc5ccccc5Cl)CC4)...</td>\n",
+       "      <td>9.92</td>\n",
+       "      <td>2014.0</td>\n",
+       "      <td>ApplicabilityDomainTutorialDataset_0017</td>\n",
+       "      <td>9.92</td>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_0025</th>\n",
        "      <td>CN(C)c1ccc(Nc2nc(NC3CCCCC3)c3[nH]cnc3n2)cc1</td>\n",
        "      <td>5.21</td>\n",
        "      <td>2005.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_0025</td>\n",
        "      <td>5.21</td>\n",
+       "      <td>True</td>\n",
        "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>ApplicabilityDomainTutorialDataset_0051</th>\n",
+       "      <td>Nc1nc(C(=O)NCc2ccccc2Cl)nc(-c2ccco2)n1</td>\n",
+       "      <td>8.31</td>\n",
+       "      <td>2009.0</td>\n",
+       "      <td>ApplicabilityDomainTutorialDataset_0051</td>\n",
+       "      <td>8.31</td>\n",
        "      <td>True</td>\n",
+       "      <td>False</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ApplicabilityDomainTutorialDataset_0056</th>\n",
@@ -374,38 +355,8 @@
        "      <td>2017.0</td>\n",
        "      <td>ApplicabilityDomainTutorialDataset_0056</td>\n",
        "      <td>5.39</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>ApplicabilityDomainTutorialDataset_0076</th>\n",
-       "      <td>O=C(NC1CC1)C12CC1C(n1cnc3c(NC4CC5CCC4C5)nc(Cl)...</td>\n",
-       "      <td>6.28</td>\n",
-       "      <td>2019.0</td>\n",
-       "      <td>ApplicabilityDomainTutorialDataset_0076</td>\n",
-       "      <td>6.28</td>\n",
-       "      <td>False</td>\n",
-       "      <td>True</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>ApplicabilityDomainTutorialDataset_0080</th>\n",
-       "      <td>CCn1c(=O)c2nc(C3CCCC3)[nH]c2n(CC)c1=O</td>\n",
-       "      <td>6.80</td>\n",
-       "      <td>2019.0</td>\n",
-       "      <td>ApplicabilityDomainTutorialDataset_0080</td>\n",
-       "      <td>6.80</td>\n",
-       "      <td>False</td>\n",
        "      <td>True</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>ApplicabilityDomainTutorialDataset_0124</th>\n",
-       "      <td>Nc1ncnc2c1ncn2C1C(O)C(O)C2(CO)CC12</td>\n",
-       "      <td>4.65</td>\n",
-       "      <td>2000.0</td>\n",
-       "      <td>ApplicabilityDomainTutorialDataset_0124</td>\n",
-       "      <td>4.65</td>\n",
        "      <td>False</td>\n",
-       "      <td>True</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
@@ -414,46 +365,46 @@
       "text/plain": [
        "                                                                                    SMILES  \\\n",
        "QSPRID                                                                                       \n",
+       "ApplicabilityDomainTutorialDataset_0006                Nc1nc(-c2ccccc2)cn2cc(-c3ccco3)nc12   \n",
+       "ApplicabilityDomainTutorialDataset_0017  Cn1c(=O)c2nc(-c3ccc(OCCCN4CCN(Cc5ccccc5Cl)CC4)...   \n",
        "ApplicabilityDomainTutorialDataset_0025        CN(C)c1ccc(Nc2nc(NC3CCCCC3)c3[nH]cnc3n2)cc1   \n",
+       "ApplicabilityDomainTutorialDataset_0051             Nc1nc(C(=O)NCc2ccccc2Cl)nc(-c2ccco2)n1   \n",
        "ApplicabilityDomainTutorialDataset_0056    O=c1[nH]c(-c2cccc(O)c2)cn2c(=O)n(-c3ccccc3)nc12   \n",
-       "ApplicabilityDomainTutorialDataset_0076  O=C(NC1CC1)C12CC1C(n1cnc3c(NC4CC5CCC4C5)nc(Cl)...   \n",
-       "ApplicabilityDomainTutorialDataset_0080              CCn1c(=O)c2nc(C3CCCC3)[nH]c2n(CC)c1=O   \n",
-       "ApplicabilityDomainTutorialDataset_0124                 Nc1ncnc2c1ncn2C1C(O)C(O)C2(CO)CC12   \n",
        "\n",
        "                                         pchembl_value_Mean    Year  \\\n",
        "QSPRID                                                                \n",
+       "ApplicabilityDomainTutorialDataset_0006                6.28  2017.0   \n",
+       "ApplicabilityDomainTutorialDataset_0017                9.92  2014.0   \n",
        "ApplicabilityDomainTutorialDataset_0025                5.21  2005.0   \n",
+       "ApplicabilityDomainTutorialDataset_0051                8.31  2009.0   \n",
        "ApplicabilityDomainTutorialDataset_0056                5.39  2017.0   \n",
-       "ApplicabilityDomainTutorialDataset_0076                6.28  2019.0   \n",
-       "ApplicabilityDomainTutorialDataset_0080                6.80  2019.0   \n",
-       "ApplicabilityDomainTutorialDataset_0124                4.65  2000.0   \n",
        "\n",
        "                                                                          QSPRID  \\\n",
        "QSPRID                                                                             \n",
+       "ApplicabilityDomainTutorialDataset_0006  ApplicabilityDomainTutorialDataset_0006   \n",
+       "ApplicabilityDomainTutorialDataset_0017  ApplicabilityDomainTutorialDataset_0017   \n",
        "ApplicabilityDomainTutorialDataset_0025  ApplicabilityDomainTutorialDataset_0025   \n",
+       "ApplicabilityDomainTutorialDataset_0051  ApplicabilityDomainTutorialDataset_0051   \n",
        "ApplicabilityDomainTutorialDataset_0056  ApplicabilityDomainTutorialDataset_0056   \n",
-       "ApplicabilityDomainTutorialDataset_0076  ApplicabilityDomainTutorialDataset_0076   \n",
-       "ApplicabilityDomainTutorialDataset_0080  ApplicabilityDomainTutorialDataset_0080   \n",
-       "ApplicabilityDomainTutorialDataset_0124  ApplicabilityDomainTutorialDataset_0124   \n",
        "\n",
        "                                         pchembl_value_Mean_original  \\\n",
        "QSPRID                                                                 \n",
+       "ApplicabilityDomainTutorialDataset_0006                         6.28   \n",
+       "ApplicabilityDomainTutorialDataset_0017                         9.92   \n",
        "ApplicabilityDomainTutorialDataset_0025                         5.21   \n",
+       "ApplicabilityDomainTutorialDataset_0051                         8.31   \n",
        "ApplicabilityDomainTutorialDataset_0056                         5.39   \n",
-       "ApplicabilityDomainTutorialDataset_0076                         6.28   \n",
-       "ApplicabilityDomainTutorialDataset_0080                         6.80   \n",
-       "ApplicabilityDomainTutorialDataset_0124                         4.65   \n",
        "\n",
-       "                                         Split_IsTrain  Split_IsOutlier  \n",
+       "                                         Split_IsOutlier  Split_IsTrain  \n",
        "QSPRID                                                                   \n",
-       "ApplicabilityDomainTutorialDataset_0025          False             True  \n",
-       "ApplicabilityDomainTutorialDataset_0056          False             True  \n",
-       "ApplicabilityDomainTutorialDataset_0076          False             True  \n",
-       "ApplicabilityDomainTutorialDataset_0080          False             True  \n",
-       "ApplicabilityDomainTutorialDataset_0124          False             True  "
+       "ApplicabilityDomainTutorialDataset_0006             True          False  \n",
+       "ApplicabilityDomainTutorialDataset_0017             True          False  \n",
+       "ApplicabilityDomainTutorialDataset_0025             True          False  \n",
+       "ApplicabilityDomainTutorialDataset_0051             True          False  \n",
+       "ApplicabilityDomainTutorialDataset_0056             True          False  "
       ]
      },
-     "execution_count": 4,
+     "execution_count": 3,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -464,7 +415,7 @@
     "from mlchemad.applicability_domains import KNNApplicabilityDomain\n",
     "\n",
     "# define applicability domain\n",
-    "applicability_domain = KNNApplicabilityDomain(scaling=None, dist=\"rogerstanimoto\")\n",
+    "applicability_domain = KNNApplicabilityDomain(scaling=None, dist=\"jaccard\", hard_threshold=0.3, k=1)\n",
     "\n",
     "# calculate compound features and split dataset into train and test\n",
     "dataset.prepareDataset(\n",
@@ -494,7 +445,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -502,10 +453,10 @@
      "output_type": "stream",
      "text": [
       "Number of compounds in train set: 3265\n",
-      "Number of compounds in test set: 703\n",
-      "Total number of compounds in train and test: 3968\n",
+      "Number of compounds in test set: 543\n",
+      "Total number of compounds in train and test: 3808\n",
       "Total number of compounds in whole dataset: 4082\n",
-      "Number removed compounds: 114\n"
+      "Number removed compounds: 274\n"
      ]
     }
    ],
@@ -535,7 +486,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -544,7 +495,7 @@
        "(4082, 2048)"
       ]
      },
-     "execution_count": 6,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -562,7 +513,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -571,13 +522,13 @@
        "<Axes: >"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -618,9 +569,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 7,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "qsprpred - WARNING - Random state supplied, but alg <class 'sklearn.neighbors._regression.KNeighborsRegressor'> does not support it. Ignoring this setting.\n"
+     ]
+    }
+   ],
    "source": [
     "from qsprpred.models import SklearnModel\n",
     "from sklearn.neighbors import KNeighborsRegressor\n",
@@ -640,7 +599,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -648,13 +607,13 @@
       "text/plain": [
        "(array([[6.336],\n",
        "        [8.325],\n",
-       "        [4.912]]),\n",
+       "        [7.21 ]]),\n",
        " array([[ True],\n",
        "        [ True],\n",
        "        [False]]))"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -662,7 +621,7 @@
    "source": [
     "smiles_list = [\"O=C(Nc1nc2ccccc2n2c(=O)n(-c3ccccc3)nc12)c1ccccc1\",\n",
     "               \"Nc1nc(-c2ccco2)c2ncn(C(=O)NCCc3ccccc3)c2n1\",\n",
-    "               \"Nc1nc(CSc2nnc(N)s2)nc(Nc2ccc(Cl)cc2)n1\"]\n",
+    "               \"CC(=O)NC1=CC=C(C=C1)O\"]\n",
     "\n",
     "model.predictMols(smiles_list, use_applicability_domain=True)"
    ]
@@ -684,7 +643,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.13"
+   "version": "3.11.5"
   }
  },
  "nbformat": 4,