feat: allow protein tokenizer to be pickled (#166)

* feat: allow protein tokenizer to be pickled

* chore: version bump

* ci: remove old python

* ci: remove old python

* fix: remove deprecated nan

* ci: remove black

* test: remove pubchem inconsistent mol

* ci: update linux for PyPI release

Author: jannisborn

.github/workflows/build.yml

@@ -17,7 +17,7 @@ jobs:
           - os: ubuntu-latest
             pip_cache_path: ~/.cache/pip
             experimental: false
-        python-version: ["3.7", "3.8", "3.9", "3.10", "3.11", "3.12"]
+        python-version: ["3.10", "3.11", "3.12"]
     defaults:
       run:
         shell: bash -l {0}  # For conda
@@ -30,14 +30,14 @@ jobs:
         uses: actions/checkout@v2
 
       - name: Cache conda
-        uses: actions/cache@v2
+        uses: actions/cache@v4
         with:
           path: ~/conda_pkgs_dir  # from: conda-incubator/setup-miniconda@v2
           key: ${{ runner.os }}-conda-${{ matrix.python-version }}-${{ env.CACHE_NUMBER }}-${{
             hashFiles('conda.yml') }}
 
       - name: Cache pip
-        uses: actions/cache@v2
+        uses: actions/cache@v4
         with:
           path: ${{ matrix.pip_cache_path }}
           key: ${{ runner.os }}-pip-${{ matrix.python-version }}-${{ env.CACHE_NUMBER }}-${{

.github/workflows/pypi.yml

@@ -6,7 +6,7 @@ on:
 
 jobs:
   deploy:
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v2
       - uses: actions/setup-python@v2

.github/workflows/style.yml

@@ -39,31 +39,3 @@ jobs:
           SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 
-  black-check:
-    name: Black
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v2
-
-      - name: Get python
-        uses: actions/setup-python@v2
-        with:
-          python-version: "3.7"
-
-      - name: Install black
-        run: pip install black==22.3.0
-
-      - name: Format
-        run: black . --check --diff --color
-
-      - name: Send Slack notification
-        uses: 8398a7/action-slack@v2
-        if: failure()
-        with:
-          status: ${{ job.status }}
-          text: ":black_circle: ${{ github.job }} :black_circle:"
-          author_name: ${{ github.actor }}
-        env:
-          SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/suggest.yml

@@ -12,7 +12,7 @@ jobs:
       - name: Get python
         uses: actions/setup-python@v2
         with:
-          python-version: "3.7"
+          python-version: "3.10"
 
       - name: Install black
         run: pip install black==22.3.0

pytoda/__init__.py

@@ -1,2 +1,2 @@
-name = 'pytoda'
-__version__ = '1.1.6'
+name = "pytoda"
+__version__ = "1.1.7"

pytoda/datasets/_csv_statistics.py

@@ -1,4 +1,5 @@
 """Abstract implementation of _CsvStatistics."""
+
 import copy
 from functools import reduce
 
@@ -36,13 +37,11 @@ def __init__(
         self.min_max_scaler = MinMaxScaler()
         self.standardizer = StandardScaler()
         self.kwargs = copy.deepcopy(kwargs)
-        self.kwargs['dtype'] = pandas_dtype
+        self.kwargs["dtype"] = pandas_dtype
 
         self.preprocess_df = (
             # may be applied to many chunks, so logic is determined once here
-            self._reindex
-            if self.initial_feature_list
-            else self._id
+            self._reindex if self.initial_feature_list else self._id
         )
         self.setup_datasource()
 
@@ -54,7 +53,7 @@ def __init__(
 
     def _reindex(self, df: pd.DataFrame) -> pd.DataFrame:
         """Ensure given order of features, creating NaN columns for missing."""
-        return df.reindex(columns=self.initial_feature_list, fill_value=np.NaN)
+        return df.reindex(columns=self.initial_feature_list, fill_value=np.nan)
 
     def _id(self, df: pd.DataFrame) -> pd.DataFrame:
         return df

pytoda/preprocessing/tests/test_crawlers.py

@@ -16,10 +16,9 @@ def test_get_smiles_from_pubchem(self) -> None:
         """Test get_smiles_from_pubchem"""
 
         for sanitize in [True, False]:
-
             # Test text mode
-            ground_truth = 'C1=CC(=CC=C1/C=C/C(=O)C2=C(C=C(C=C2)O)O)O'
-            for query, drug in zip(['name', 'cid'], ['isoliquiritigenin', 638278]):
+            ground_truth = "C1=CC(=CC=C1/C=C/C(=O)C2=C(C=C(C=C2)O)O)O"
+            for query, drug in zip(["name", "cid"], ["isoliquiritigenin", 638278]):
                 smiles = get_smiles_from_pubchem(
                     drug,
                     use_isomeric=True,
@@ -29,8 +28,8 @@ def test_get_smiles_from_pubchem(self) -> None:
                 )
             self.assertEqual(smiles, ground_truth)
 
-            ground_truth = 'C1=CC(=CC=C1C=CC(=O)C2=C(C=C(C=C2)O)O)O'
-            for query, drug in zip(['name', 'cid'], ['isoliquiritigenin', 638278]):
+            ground_truth = "C1=CC(=CC=C1C=CC(=O)C2=C(C=C(C=C2)O)O)O"
+            for query, drug in zip(["name", "cid"], ["isoliquiritigenin", 638278]):
                 smiles = get_smiles_from_pubchem(
                     drug,
                     use_isomeric=False,
@@ -39,9 +38,9 @@ def test_get_smiles_from_pubchem(self) -> None:
                     query_type=query,
                 )
             # mac-os irreproducible stochastic failure on ubuntu
-            self.assertIn(smiles, [ground_truth, ''])
+            self.assertIn(smiles, [ground_truth, ""])
 
-            drug = 'isoliquiritigenin'
+            drug = "isoliquiritigenin"
             if not sanitize:
                 with self.assertRaises(ValueError):
                     get_smiles_from_pubchem(
@@ -52,48 +51,52 @@ def test_get_smiles_from_pubchem(self) -> None:
                         drug, use_isomeric=False, kekulize=False, sanitize=sanitize
                     )
             else:
-                ground_truth = 'O=C(/C=C/c1ccc(O)cc1)c1ccc(O)cc1O'
+                ground_truth = "O=C(/C=C/c1ccc(O)cc1)c1ccc(O)cc1O"
 
                 smiles = get_smiles_from_pubchem(
                     drug, use_isomeric=True, kekulize=False, sanitize=sanitize
                 )
                 self.assertEqual(smiles, ground_truth)
 
-                ground_truth = 'O=C(C=Cc1ccc(O)cc1)c1ccc(O)cc1O'
+                ground_truth = "O=C(C=Cc1ccc(O)cc1)c1ccc(O)cc1O"
                 smiles = get_smiles_from_pubchem(
                     drug, use_isomeric=False, kekulize=False, sanitize=sanitize
                 )
                 self.assertEqual(smiles, ground_truth)
 
         # Test molecule where landing page has several entries
         gt_smiles = (
-            'CC12C(C(CC(O1)N3C4=CC=CC=C4C5=C6C(=C7C8=CC=CC=C8N2C7=C53)CNC6=O)NC)OC'
+            "CC12C(C(CC(O1)N3C4=CC=CC=C4C5=C6C(=C7C8=CC=CC=C8N2C7=C53)CNC6=O)NC)OC"
         )
-        drug = 'Staurosporine'
+        drug = "Staurosporine"
         smiles = get_smiles_from_pubchem(drug, use_isomeric=False, kekulize=True)
         self.assertEqual(smiles, gt_smiles)
 
     def test_query_pubchem(self) -> None:
         """Test query_pubchem"""
         smiles_list = [
-            'O1C=CC=NC(=O)C1=O',
-            'CC(N)S(O)(=O)C(C)CC(C(C)C)c1cc(F)cc(F)c1',
-            'Clc1ccccc2ccnc12',
+            # "O1C=CC=NC(=O)C1=O",
+            "CC(N)S(O)(=O)C(C)CC(C(C)C)c1cc(F)cc(F)c1",
+            "Clc1ccccc2ccnc12",
+        ]
+        ground_truths = [
+            # (True, 67945516),
+            (False, -2),
+            (False, -1),
         ]
-        ground_truths = [(True, 67945516), (False, -2), (False, -1)]
         for gt, smiles in zip(ground_truths, smiles_list):
             self.assertTupleEqual(query_pubchem(smiles), gt)
 
     def test_remove_pubchem_smiles(self) -> None:
         """Test remove_pubchem_smiles"""
         smiles_list = [
-            'O1C=CC=NC(=O)C1=O',
-            'CC(N)S(O)(=O)C(C)CC(C(C)C)c1cc(F)cc(F)c1',
-            'Clc1ccccc2ccnc12',
+            "O1C=CC=NC(=O)C1=O",
+            "CC(N)S(O)(=O)C(C)CC(C(C)C)c1cc(F)cc(F)c1",
+            "Clc1ccccc2ccnc12",
         ]
-        ground_truth = ['CC(N)S(O)(=O)C(C)CC(C(C)C)c1cc(F)cc(F)c1', 'Clc1ccccc2ccnc12']
+        ground_truth = ["CC(N)S(O)(=O)C(C)CC(C(C)C)c1cc(F)cc(F)c1", "Clc1ccccc2ccnc12"]
         self.assertListEqual(remove_pubchem_smiles(smiles_list), ground_truth)
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     unittest.main()