qsprpred.data.chem.standardizers package

Submodules

qsprpred.data.chem.standardizers.base module

exception qsprpred.data.chem.standardizers.base.ChemStandardizationException[source]

Bases: Exception

Exception raised when standardization fails.

add_note(note, /): Add a note to the exception

args

with_traceback(tb, /): Set self.__traceback__ to tb and return self.

class qsprpred.data.chem.standardizers.base.ChemStandardizer[source]

Bases: ABC

Standardizer to convert SMILES to a standardized form.

This class defines an interface of a uniquely identifiable standardizer. The getID method should return a unique identifier for the standardizer based on its settings. Standardizes that have the same ID should produce the same standardized form for a given SMILES.

The main method of the class is convertSMILES, which should convert a given SMILES to a standardized form based on the settings of the standardizer.

abstract convertSMILES(smiles: str) → str | None[source]

Convert the SMILES to a standardized form.

Parameters:: smiles (str) – SMILES to be converted
Returns:: The standardized SMILES string or None if standardization fails or the molecule is deemed invalid.
Return type:: str | None
Raises:: ChemStandardizationException – if standardization fails, but the upstream code should be notified and handle the exception.

abstract classmethod fromSettings(settings: dict) → ChemStandardizer[source]: Create a new standardizer from a settings dictionary.

classmethod fromSettingsFile(path: str) → ChemStandardizer[source]

Load the standardizer from a settings file in JSON format.

Parameters:: path (str) – Path to the settings file.
Returns:: The standardizer loaded from the settings file.
Return type:: ChemStandardizer

getHashID() → str[source]

Get the hash ID of the standardizer. This is simply the MD5 hash of the unique identifier of the standardizer.

Returns:: The hash ID of the standardizer
Return type:: str

abstract getID() → str[source]

Return the unique identifier of the standardizer. This method should return a unique identifier based on the settings of the standardizer.

Two standardizers with the same settings should have the same ID and produce the same standardized form for a given SMILES.

Returns:: The unique identifier of the standardizer.
Return type:: str

abstract property settings: dict: Settings of the standardizer. It should contain complete information needed to initialize another equivalent standardizer.

class qsprpred.data.chem.standardizers.base.Standardizable[source]

Bases: ABC

Interface for objects that use chemical standardization with ` ChemStandardizer objects.

abstract applyStandardizer(standardizer: ChemStandardizer)[source]

Apply a standardizer to the SMILES in the store.

Parameters:: standardizer (ChemStandardizer) – The standardizer to apply

abstract property standardizer: ChemStandardizer

Get the standardizer used by the store.

Returns:: The standardizer used by the store.
Return type:: ChemStandardizer

qsprpred.data.chem.standardizers.check_smiles module

class qsprpred.data.chem.standardizers.check_smiles.CheckSmilesValid(id_prop: str | None = 'ID')[source]

Bases: MolProcessorWithID

Processor to check the validity of the SMILES.

Initialize the processor with the name of the property that contains the molecule’s unique identifier.

Parameters:: id_prop (str) – Name of the property that contains the molecule’s unique identifier. Defaults to “QSPRID”.

iterMolsAndIDs(mols, props: dict[str, list] | None)

Iterate over molecules and their corresponding IDs regardless of the input molecule format. This is just a helper function that will detect the input and yield the molecule and its ID.

Parameters:

mols (list[str | Mol | StoredMol]) – A list of SMILES or RDKit molecules to process.
props (dict) – An optional dictionary of properties related to the molecules to process.

Returns:

A tuple of the molecules and their IDs.

Return type:

tuple[Mol, str]

property requiredProps: list[str]: The properties required by the processor. This is to inform the caller that the processor requires certain properties to be passed to the __call__ method or via the props attribute of StoredMol instances.

property supportsParallel: bool: Return True if the processor supports parallel processing.

class qsprpred.data.chem.standardizers.check_smiles.ValidationStandardizer[source]

Bases: ChemStandardizer

Standardizer that checks the validity of the SMILES by attempting to sanitize the molecule using RDKit.

Variables:: checker (CheckSmilesValid) – Processor to check the validity of the SMILES

Initialize the standardizer.

Raises:: ValueError – If the SMILES is invalid

convertSMILES(smiles: str) → str | None[source]

Check the validity of the SMILES.

Parameters:: smiles (str) – SMILES to be checked
Returns:: the standardized SMILES
Return type:: str | None

classmethod fromSettings(settings: dict) → ValidationStandardizer[source]

Create a standardizer from settings. In this case, the settings are ignored.

Parameters:: settings (dict) – Settings of the standardizer
Returns:: The standardizer created from settings
Return type:: ValidationStandardizer

classmethod fromSettingsFile(path: str) → ChemStandardizer

Load the standardizer from a settings file in JSON format.

Parameters:: path (str) – Path to the settings file.
Returns:: The standardizer loaded from the settings file.
Return type:: ChemStandardizer

getHashID() → str

Get the hash ID of the standardizer. This is simply the MD5 hash of the unique identifier of the standardizer.

Returns:: The hash ID of the standardizer
Return type:: str

getID()[source]: Return the unique identifier of the standardizer. In this case, it is just “ValidationStandardizer”. There are no settings to consider.

property settings: Settings of the standardizer. Empty in this case since there is nothing to set except the default settings.

qsprpred.data.chem.standardizers.chembl module

class qsprpred.data.chem.standardizers.chembl.ChemblStandardizer(isomeric_smiles: bool = True, sanitize: bool = True)[source]

Bases: ChemStandardizer

Standardizer using the ChEMBL standardizer.

Variables:

isomericSmiles (bool) – return the isomeric smiles.
sanitize (bool) – sanitize SMILES before standardization.

Initialize the ChEMBL standardizer.

Parameters:

isomeric_smiles (bool) – return the isomeric smiles. Defaults to True.
sanitize (bool) – sanitize SMILES before standardization. Defaults to True.

convertSMILES(smiles: str) → str | None[source]

Standardize SMILES using the ChEMBL standardizer.

Parameters:: smiles (str) – SMILES to be standardized
Returns:: standardized SMILES string or None if standardization failed.
Return type:: (str)

classmethod fromSettings(settings: dict) → ChemblStandardizer[source]

Create a standardizer from settings.

Parameters:: settings (dict) – Settings of the standardizer
Returns:: The standardizer created from settings
Return type:: (ChemblStandardizer)

classmethod fromSettingsFile(path: str) → ChemStandardizer

Load the standardizer from a settings file in JSON format.

Parameters:: path (str) – Path to the settings file.
Returns:: The standardizer loaded from the settings file.
Return type:: ChemStandardizer

getHashID() → str

Get the hash ID of the standardizer. This is simply the MD5 hash of the unique identifier of the standardizer.

Returns:: The hash ID of the standardizer
Return type:: str

getID() → str[source]

Return the unique identifier of the standardizer.

In this case, the identifier starts with “ChEMBLStandardizer” followed by the settings of the standardizer concatenated with “~”.

Returns:: unique identifier of the standardizer
Return type:: (str)

property settings: dict: Settings of the standardizer. It should contain complete information needed to initialize another equivalent standardizer.

qsprpred.data.chem.standardizers.chembl.chembl_smi_standardizer(smi: str, isomeric_smiles: bool = True, sanitize: bool = True) → str | None[source]

Standardize SMILES using ChEMBL standardizer.

Parameters:

smi (str) – SMILES string to be standardized.
isomeric_smiles (bool) – return the isomeric smiles. Defaults to True.
sanitize (bool) – applies sanitization using the ChEMBL standardizer. Defaults to True.

Returns:

standardized SMILES string or None if standardization failed.

Return type:

(str)

qsprpred.data.chem.standardizers.naive module

class qsprpred.data.chem.standardizers.naive.NaiveStandardizer[source]

Bases: ChemStandardizer

Naive standardizer

Briefly, the standardization process involves disconnecting metals, normalizing, removing salts (largest fragment) and charges. See qsprpred.data.chem.standardizers.naive.standardize_mol for more details.

convertSMILES(smiles: str) → str | None[source]

Standardize SMILES using standardize_mol.

Parameters:: smiles (str) – SMILES to be standardized
Returns:: standardized SMILES or None if SMILES could not be standardized
Return type:: str | None

classmethod fromSettings(settings: dict) → NaiveStandardizer[source]

Create a naive standardizer from settings. In this case, the settings are ignored.

Parameters:: settings (dict) – settings of the standardizer
Returns:: a naive standardizer
Return type:: NaiveStandardizer

classmethod fromSettingsFile(path: str) → ChemStandardizer

Load the standardizer from a settings file in JSON format.

Parameters:: path (str) – Path to the settings file.
Returns:: The standardizer loaded from the settings file.
Return type:: ChemStandardizer

getHashID() → str

Get the hash ID of the standardizer. This is simply the MD5 hash of the unique identifier of the standardizer.

Returns:: The hash ID of the standardizer
Return type:: str

getID() → str[source]: Return the unique identifier of the standardizer, which in this case is “NaiveStandardizer” without any settings.

property settings: dict: Settings of the standardizer. They are empty in this case.

qsprpred.data.chem.standardizers.naive.standardize_mol(mol) → str | None[source]

Standardizes SMILES and removes fragments

Standardizes SMILES using RDKit MolStandardize to disconnect metals, normalize, remove salts (largest fragment), and uncharge. Followed by a second round of disconnecting metals and normalizing. Finally, the SMILES is canonicalized.

Parameters:: mol (rdkit.Chem.rdchem.Mol) – RDKit molecule object
Returns:: Standardized SMILES or None if SMILES could not be standardized or if SMILES does not contain carbon or contains salts after standardization
Return type:: (str | None)

qsprpred.data.chem.standardizers.papyrus module

class qsprpred.data.chem.standardizers.papyrus.PapyrusStandardizer(keep_stereo: bool = True, canonize: bool = True, mixture_handling: Literal['keep_largest', 'filter', 'keep'] = 'keep_largest', remove_additional_salts: bool = True, remove_additional_metals: bool = True, filter_inorganic: bool = False, filter_non_small_molecule: bool = True, small_molecule_min_mw: float = 200, small_molecule_max_mw: float = 800, canonicalize_tautomer: bool = True, tautomer_max_tautomers: int = 4294967295, extra_organic_atoms: list | None = None, extra_metals: list | None = None, extra_salts: list | None = None, uncharge: bool = True)[source]

Bases: ChemStandardizer

Papyrus standardizer

Uses Papyrus (>v05.6) standardization protecol to standardize SMILES.

Béquignon, O.J.M., Bongers, B.J., Jespers, W. et al. Papyrus: a large-scale curated dataset aimed at bioactivity predictions. J Cheminform 15, 3 (2023). https://doi.org/10.1186/s13321-022-00672-x

Variables:: settings (dict) – Settings of the standardizer

Initialize Papyrus standardizer

Parameters:

keep_stereo (bool, optional) – Keep stereochemistry.
canonize (bool, optional) – Canonicalize SMILES.
mixture_handling (Literal["keep_largest", "filter", "keep"], optional) – How to handle mixtures. Defaults to “keep_largest”.
remove_additional_salts (bool, optional) – Removes a custom set of fragments if present in the molecule object.
remove_additional_metals (bool, optional) – Removes metal fragments if present in the molecule object. Ignored if remove_additional_salts is set to False.
filter_inorganic (bool, optional) – Filter inorganic molecules.
filter_non_small_molecule (bool, optional) – Filter non-small molecules.
small_molecule_min_mw (float, optional) – Minimum molecular weight of small molecules.
small_molecule_max_mw (float, optional) – Maximum molecular weight of small molecules.
canonicalize_tautomer (bool, optional) – Canonicalize tautomers.
tautomer_max_tautomers (int, optional) – Maximum number of tautomers to consider by the tautomer search algorithm (<2^32).
extra_organic_atoms (list, optional) – Extra organic atoms to consider in addition to the default set (Papyrus_standardizer.ORGANIC_ATOMS).
extra_metals (list, optional) – Extra metals to consider in addition to the default set (Papyrus_standardizer.METALS).
extra_salts (list, optional) – Extra salts to consider in addition to the default set (Papyrus_standardizer.SALTS).
uncharge (bool, optional) – Uncharge molecules.

convertSMILES(smiles: str, verbose: bool = False) → str | None[source]

Standardize SMILES using Papyrus standardization protocol.

Parameters:

smiles (str) – SMILES to be standardized
verbose (bool, optional) – Print verbose output. Defaults to False.

Returns:

a tuple where the first element is the standardized SMILES and the second element is the original SMILES

Return type:

tuple[str | None, str]

fromSettings(settings: dict) → PapyrusStandardizer[source]

Create a Papyrus standardizer from settings.

Parameters:: settings (dict) – settings of the standardizer
Returns:: a Papyrus standardizer
Return type:: PapyrusStandardizer

classmethod fromSettingsFile(path: str) → ChemStandardizer

Load the standardizer from a settings file in JSON format.

Parameters:: path (str) – Path to the settings file.
Returns:: The standardizer loaded from the settings file.
Return type:: ChemStandardizer

getHashID() → str

Get the hash ID of the standardizer. This is simply the MD5 hash of the unique identifier of the standardizer.

Returns:: The hash ID of the standardizer
Return type:: str

getID() → str[source]

Get the ID of the standardizer.

In this case, the ID is based on the settings of the standardizer. It starts with ‘PapyrusStandardizer’ followed by a tilde and the settings concatenated with a colon.

Returns:: ID of the standardizer
Return type:: str

property settings: dict: Settings of the standardizer. It should contain complete information needed to initialize another equivalent standardizer.

Module contents

class qsprpred.data.chem.standardizers.ChemStandardizer[source]

Bases: ABC

Standardizer to convert SMILES to a standardized form.

This class defines an interface of a uniquely identifiable standardizer. The getID method should return a unique identifier for the standardizer based on its settings. Standardizes that have the same ID should produce the same standardized form for a given SMILES.

The main method of the class is convertSMILES, which should convert a given SMILES to a standardized form based on the settings of the standardizer.

abstract convertSMILES(smiles: str) → str | None[source]

Convert the SMILES to a standardized form.

Parameters:: smiles (str) – SMILES to be converted
Returns:: The standardized SMILES string or None if standardization fails or the molecule is deemed invalid.
Return type:: str | None
Raises:: ChemStandardizationException – if standardization fails, but the upstream code should be notified and handle the exception.

abstract classmethod fromSettings(settings: dict) → ChemStandardizer[source]: Create a new standardizer from a settings dictionary.

classmethod fromSettingsFile(path: str) → ChemStandardizer[source]

Load the standardizer from a settings file in JSON format.

Parameters:: path (str) – Path to the settings file.
Returns:: The standardizer loaded from the settings file.
Return type:: ChemStandardizer

getHashID() → str[source]

Get the hash ID of the standardizer. This is simply the MD5 hash of the unique identifier of the standardizer.

Returns:: The hash ID of the standardizer
Return type:: str

abstract getID() → str[source]

Return the unique identifier of the standardizer. This method should return a unique identifier based on the settings of the standardizer.

Two standardizers with the same settings should have the same ID and produce the same standardized form for a given SMILES.

Returns:: The unique identifier of the standardizer.
Return type:: str

abstract property settings: dict: Settings of the standardizer. It should contain complete information needed to initialize another equivalent standardizer.