Merge branch 'main' into dataset_similarity

Author: JochenSiegWork

molpipeline/abstract_pipeline_elements/mol2any/mol2bitvector.py

@@ -8,6 +8,7 @@
 from molpipeline.mol2any.mol2morgan_fingerprint import MolToMorganFP
 from molpipeline.mol2any.mol2net_charge import MolToNetCharge
 from molpipeline.mol2any.mol2path_fingerprint import Mol2PathFP
+from molpipeline.mol2any.mol2pharmacophore2d_fingerprint import MolToPharmacophore2DFP
 from molpipeline.mol2any.mol2rdkit_phys_chem import MolToRDKitPhysChem
 from molpipeline.mol2any.mol2smiles import MolToSmiles
 
@@ -21,6 +22,7 @@
     "MolToMACCSFP",
     "MolToMorganFP",
     "MolToNetCharge",
+    "MolToPharmacophore2DFP",
     "MolToRDKitPhysChem",
     "MolToSmiles",
 ]

molpipeline/mol2any/__init__.py

@@ -6,6 +6,7 @@
 from rdkit.DataStructs import ExplicitBitVect
 
 from molpipeline.abstract_pipeline_elements.mol2any.mol2bitvector import (
+    FPReturnAsOption,
     MolToFingerprintPipelineElement,
 )
 from molpipeline.utils.molpipeline_types import RDKitMol
@@ -21,10 +22,44 @@ class MolToMACCSFP(MolToFingerprintPipelineElement):
     """
 
     _n_bits = 167  # MACCS keys have 166 bits + 1 bit for an all-zero vector (bit 0)
-    _feature_names = [f"maccs_{i}" for i in range(_n_bits)]
+
+    def __init__(
+        self,
+        return_as: FPReturnAsOption = "sparse",
+        name: str = "MolToMACCSFP",
+        n_jobs: int = 1,
+        uuid: str | None = None,
+    ):
+        """Initialize MolToMACCSFP.
+
+        Parameters
+        ----------
+        return_as : FPReturnAsOption, default="sparse"
+            Type of output. When "sparse" the fingerprints will be returned as a
+            scipy.sparse.csr_matrix holding a sparse representation of the bit vectors.
+            With "dense" a numpy matrix will be returned. With "rdkit" the fingerprints
+            will be returned as a list of one of RDKit's ExplicitBitVect,
+            IntSparseBitVect, UIntSparseBitVect, etc. depending on the fingerprint
+            and parameters.
+        name : str, default="MolToMACCSFP"
+            Name of PipelineElement.
+        n_jobs : int, default=1
+            Number of cores to use.
+        uuid : str | None, optional
+            UUID of the PipelineElement.
+
+        """
+        super().__init__(  # pylint: disable=R0801
+            return_as=return_as,
+            name=name,
+            n_jobs=n_jobs,
+            uuid=uuid,
+        )
+        self._feature_names = [f"maccs_{i}" for i in range(self._n_bits)]
 
     def pretransform_single(
-        self, value: RDKitMol
+        self,
+        value: RDKitMol,
     ) -> dict[int, int] | npt.NDArray[np.int_] | ExplicitBitVect:
         """Transform a single molecule to MACCS key fingerprint.
 
@@ -45,7 +80,7 @@ def pretransform_single(
 
         """
         fingerprint = MACCSkeys.GenMACCSKeys(value)  # type: ignore[attr-defined]
-        if self._return_as == "explicit_bit_vect":
+        if self._return_as == "rdkit":
             return fingerprint
         if self._return_as == "dense":
             return np.array(fingerprint)

molpipeline/mol2any/mol2maccs_key_fingerprint.py

@@ -2,7 +2,7 @@
 
 from __future__ import annotations  # for all the python 3.8 users out there.
 
-from typing import Any, Literal
+from typing import TYPE_CHECKING, Any
 
 try:
     from typing import Self  # type: ignore[attr-defined]
@@ -15,8 +15,11 @@
 
 from molpipeline.abstract_pipeline_elements.mol2any.mol2bitvector import (
     ABCMorganFingerprintPipelineElement,
+    FPReturnAsOption,
 )
-from molpipeline.utils.molpipeline_types import RDKitMol
+
+if TYPE_CHECKING:
+    from molpipeline.utils.molpipeline_types import RDKitMol
 
 
 class MolToMorganFP(ABCMorganFingerprintPipelineElement):
@@ -33,7 +36,7 @@ def __init__(
         use_features: bool = False,
         n_bits: int = 2048,
         counted: bool = False,
-        return_as: Literal["sparse", "dense", "explicit_bit_vect"] = "sparse",
+        return_as: FPReturnAsOption = "sparse",
         name: str = "MolToMorganFP",
         n_jobs: int = 1,
         uuid: str | None = None,
@@ -52,12 +55,13 @@ def __init__(
         counted: bool, default=False
             If True, the fingerprint will be counted.
             If False, the fingerprint will be binary.
-        return_as: Literal["sparse", "dense", "explicit_bit_vect"], default="sparse"
+        return_as: FPReturnAsOption, default="sparse"
             Type of output. When "sparse" the fingerprints will be returned as a
             scipy.sparse.csr_matrix holding a sparse representation of the bit vectors.
-            With "dense" a numpy matrix will be returned.
-            With "explicit_bit_vect" the fingerprints will be returned as a list of
-            RDKit's rdkit.DataStructs.cDataStructs.ExplicitBitVect.
+            With "dense" a numpy matrix will be returned. With "rdkit" the fingerprints
+            will be returned as a list of one of RDKit's ExplicitBitVect,
+            IntSparseBitVect, UIntSparseBitVect, etc. depending on the fingerprint
+            and parameters.
         name: str, default="MolToMorganFP"
             Name of PipelineElement
         n_jobs: int, default=1
@@ -88,7 +92,7 @@ def __init__(
         )
         if not isinstance(n_bits, int) or n_bits < 1:
             raise ValueError(
-                f"Number of bits has to be a integer > 0! (Received: {n_bits})"
+                f"Number of bits has to be a integer > 0! (Received: {n_bits})",
             )
         self._n_bits = n_bits
         self._feature_names = [f"morgan_{i}" for i in range(self._n_bits)]
@@ -105,6 +109,7 @@ def get_params(self, deep: bool = True) -> dict[str, Any]:
         -------
         dict[str, Any]
             Dictionary of parameters.
+
         """
         parameters = super().get_params(deep)
         if deep:
@@ -125,6 +130,7 @@ def set_params(self, **parameters: Any) -> Self:
         -------
         Self
             MolToMorganFP pipeline element with updated parameters.
+
         """
         parameter_copy = dict(parameters)
         n_bits = parameter_copy.pop("n_bits", None)
@@ -143,6 +149,7 @@ def _get_fp_generator(
         -------
         rdFingerprintGenerator.FingerprintGenerator
             RDKit fingerprint generator.
+
         """
         return rdFingerprintGenerator.GetMorganGenerator(
             radius=self.radius,
@@ -160,12 +167,13 @@ def _explain_rdmol(self, mol_obj: RDKitMol) -> dict[int, list[tuple[int, int]]]:
         Returns
         -------
         dict[int, list[tuple[int, int]]]
-            Dictionary with bit position as key and list of tuples with atom index and radius as value.
+            Dictionary with bit position as key and list of tuples with atom index
+            and radius as value.
+
         """
         fp_generator = self._get_fp_generator()
         additional_output = AllChem.AdditionalOutput()
         additional_output.AllocateBitInfoMap()
         # using the dense fingerprint here, to get indices after folding
         _ = fp_generator.GetFingerprint(mol_obj, additionalOutput=additional_output)
-        bit_info = additional_output.GetBitInfoMap()
-        return bit_info
+        return additional_output.GetBitInfoMap()

molpipeline/mol2any/mol2morgan_fingerprint.py

@@ -2,7 +2,7 @@
 
 from __future__ import annotations  # for all the python 3.8 users out there.
 
-from typing import Any, Literal
+from typing import Any
 
 try:
     from typing import Self  # type: ignore[attr-defined]
@@ -14,6 +14,7 @@
 from rdkit.Chem import rdFingerprintGenerator
 
 from molpipeline.abstract_pipeline_elements.mol2any.mol2bitvector import (
+    FPReturnAsOption,
     MolToRDKitGenFPElement,
 )
 
@@ -39,7 +40,7 @@ def __init__(
         num_bits_per_feature: int = 2,
         atom_invariants_generator: Any = None,
         counted: bool = False,
-        return_as: Literal["sparse", "dense", "explicit_bit_vect"] = "sparse",
+        return_as: FPReturnAsOption = "sparse",
         name: str = "Mol2PathFP",
         n_jobs: int = 1,
         uuid: str | None = None,
@@ -71,12 +72,13 @@ def __init__(
         counted: bool, default=False
             If True, the fingerprint will be counted.
             If False, the fingerprint will be binary.
-        return_as: Literal["sparse", "dense", "explicit_bit_vect"], default="sparse"
+        return_as: FPReturnAsOption, default="sparse"
             Type of output. When "sparse" the fingerprints will be returned as a
             scipy.sparse.csr_matrix holding a sparse representation of the bit vectors.
-            With "dense" a numpy matrix will be returned.
-            With "explicit_bit_vect" the fingerprints will be returned as a list of
-            RDKit's rdkit.DataStructs.cDataStructs.ExplicitBitVect.
+            With "dense" a numpy matrix will be returned. With "rdkit" the fingerprints
+            will be returned as a list of one of RDKit's ExplicitBitVect,
+            IntSparseBitVect, UIntSparseBitVect, etc. depending on the fingerprint
+            and parameters.
         name: str, default="MolToMorganFP"
             Name of PipelineElement
         n_jobs: int, default=1