skchem.io package¶

Submodules¶

skchem.io.sdf module¶

# skchem.io.sdf

Defining input and output operations for sdf files.

skchem.io.sdf.read_sdf(sdf, error_bad_mol=False, warn_bad_mol=True, nmols=None, skipmols=None, skipfooter=None, read_props=True, mol_props=False, *args, **kwargs)[source]¶

Read an sdf file into a pd.DataFrame.

The function wraps the RDKit ForwardSDMolSupplier object.

Parameters:	sdf (str or file-like) – The location of data to load as a file path, or a file-like object. error_bad_mol (bool) – Whether an error should be raised if a molecule fails to parse. Default is False. warn_bad_mol (bool) – Whether a warning should be output if a molecule fails to parse. Default is True. nmols (int) – The number of molecules to read. If None, read all molecules. Default is None. skipmols (int) – The number of molecules to skip at start. Default is 0. skipfooter (int) – The number of molecules to skip from the end. Default is 0. read_props (bool) – Whether to read the properties into the data frame. Default is True. mol_props (bool) – Whether to keep properties in the molecule dictionary after they are extracted to the DataFrame. Default is False. kwargs (args,) – Arguments will be passed to RDKit ForwardSDMolSupplier.
Returns:	The loaded data frame, with Mols supplied in the structure field.
Return type:	pandas.DataFrame

See also

rdkit.Chem.SDForwardMolSupplier skchem.read_smiles

skchem.io.sdf.write_sdf(data, sdf, write_cols=True, index_as_name=True, mol_props=False, *args, **kwargs)[source]¶

Write an sdf file from a dataframe.

Parameters:

data (pandas.Series or pandas.DataFrame) – Pandas data structure with a structure column containing compounds to serialize. sdf (str or file-like) – A file path or file-like object specifying where to write the compound data. write_cols (bool) – Whether columns should be written as props. Default True. index_as_name (bool) – Whether to use index as the header, or the molecule’s name. Default is True. mol_props (bool) – Whether to write properties in the Mol dictionary in addition to fields in the frame.

Warn:

This function will change the names of the compounds if the index_as_name argument is True, and will delete all properties in the molecule dictionary if mol_props is False.

skchem.io.smiles module¶

# skchem.io.smiles

Defining input and output operations for smiles files.

skchem.io.smiles.read_smiles(smiles_file, smiles_column=0, name_column=None, delimiter='\t', title_line=False, error_bad_mol=False, warn_bad_mol=True, drop_bad_mol=True, *args, **kwargs)[source]¶

Read a smiles file into a pandas dataframe.

The class wraps the pandas read_csv function.

smiles_file (str, file-like):

Location of data to load, specified as a string or passed directly as a file-like object. URLs may also be used, see the pandas.read_csv documentation.

smiles_column (int):

The column index at which SMILES are provided. Defaults to 0.

name_column (int):

The column index at which compound names are provided, for use as the index in the DataFrame. If None, use the default index. Defaults to None.

delimiter (str):

The delimiter used. Defaults to t.

title_line (bool):

Whether a title line is provided, to use as column titles. Defaults to False.

error_bad_mol (bool):

Whether an error should be raised when a molecule fails to parse. Defaults to False.

warn_bad_mol (bool):

Whether a warning should be raised when a molecule fails to parse. Defaults to True.

drop_bad_mol (bool):

If true, drop any column with smiles that failed to parse. Otherwise, the field is None. Defaults to True.

args, kwargs:

Arguments will be passed to pandas read_csv arguments.

Returns:	The loaded data frame, with Mols supplied in the structure field.
Return type:	pandas.DataFrame

See also

pandas.read_csv skchem.Mol.from_smiles skchem.io.sdf

skchem.io.smiles.write_smiles(data, smiles_path)[source]¶

Write a dataframe to a smiles file.

Parameters:	data (pd.Series or pd.DataFrame) – The dataframe to write. smiles_path (str) – The path to write the dataframe to.

Module contents¶

skchem.io

Module defining input and output methods in scikit-chem.

skchem.io.read_sdf(sdf, error_bad_mol=False, warn_bad_mol=True, nmols=None, skipmols=None, skipfooter=None, read_props=True, mol_props=False, *args, **kwargs)[source]¶

Read an sdf file into a pd.DataFrame.

The function wraps the RDKit ForwardSDMolSupplier object.

Parameters:	sdf (str or file-like) – The location of data to load as a file path, or a file-like object. error_bad_mol (bool) – Whether an error should be raised if a molecule fails to parse. Default is False. warn_bad_mol (bool) – Whether a warning should be output if a molecule fails to parse. Default is True. nmols (int) – The number of molecules to read. If None, read all molecules. Default is None. skipmols (int) – The number of molecules to skip at start. Default is 0. skipfooter (int) – The number of molecules to skip from the end. Default is 0. read_props (bool) – Whether to read the properties into the data frame. Default is True. mol_props (bool) – Whether to keep properties in the molecule dictionary after they are extracted to the DataFrame. Default is False. kwargs (args,) – Arguments will be passed to RDKit ForwardSDMolSupplier.
Returns:	The loaded data frame, with Mols supplied in the structure field.
Return type:	pandas.DataFrame

See also

rdkit.Chem.SDForwardMolSupplier skchem.read_smiles

skchem.io.write_sdf(data, sdf, write_cols=True, index_as_name=True, mol_props=False, *args, **kwargs)[source]¶

Write an sdf file from a dataframe.

Parameters:

data (pandas.Series or pandas.DataFrame) – Pandas data structure with a structure column containing compounds to serialize. sdf (str or file-like) – A file path or file-like object specifying where to write the compound data. write_cols (bool) – Whether columns should be written as props. Default True. index_as_name (bool) – Whether to use index as the header, or the molecule’s name. Default is True. mol_props (bool) – Whether to write properties in the Mol dictionary in addition to fields in the frame.

Warn:

This function will change the names of the compounds if the index_as_name argument is True, and will delete all properties in the molecule dictionary if mol_props is False.

skchem.io.read_smiles(smiles_file, smiles_column=0, name_column=None, delimiter='\t', title_line=False, error_bad_mol=False, warn_bad_mol=True, drop_bad_mol=True, *args, **kwargs)[source]¶

Read a smiles file into a pandas dataframe.

The class wraps the pandas read_csv function.

smiles_file (str, file-like):

Location of data to load, specified as a string or passed directly as a file-like object. URLs may also be used, see the pandas.read_csv documentation.

smiles_column (int):

The column index at which SMILES are provided. Defaults to 0.

name_column (int):

The column index at which compound names are provided, for use as the index in the DataFrame. If None, use the default index. Defaults to None.

delimiter (str):

The delimiter used. Defaults to t.

title_line (bool):

Whether a title line is provided, to use as column titles. Defaults to False.

error_bad_mol (bool):

Whether an error should be raised when a molecule fails to parse. Defaults to False.

warn_bad_mol (bool):

Whether a warning should be raised when a molecule fails to parse. Defaults to True.

drop_bad_mol (bool):

If true, drop any column with smiles that failed to parse. Otherwise, the field is None. Defaults to True.

args, kwargs:

Arguments will be passed to pandas read_csv arguments.

Returns:	The loaded data frame, with Mols supplied in the structure field.
Return type:	pandas.DataFrame

See also

pandas.read_csv skchem.Mol.from_smiles skchem.io.sdf

skchem.io.write_smiles(data, smiles_path)[source]¶

Write a dataframe to a smiles file.

Parameters:	data (pd.Series or pd.DataFrame) – The dataframe to write. smiles_path (str) – The path to write the dataframe to.

skchem.io.read_config(conf)[source]¶

Deserialize an object from a config dict.

Parameters:	conf (dict) – The config dict to deseriailize.
Returns:	object

Note

config is different from params, in that it specifies the class. The params dict is a subdict in config.

skchem.io.write_config(obj)[source]¶

Serialize an object to a config dict.

skchem.io.read_yaml(conf)[source]¶

Deserialize an object from a yaml file, filename or str.

Parameters:	yaml (str or filelike) – The yaml file to deserialize.
Returns:	object

skchem.io.write_yaml(obj, target=None)[source]¶

Serialize a scikit-chem object to yaml.

skchem.io.read_json(conf)[source]¶

Deserialize an object from a json file, filename or str.

Parameters:	json (str or filelike) – The json file to deserialize.
Returns:	object

skchem.io.write_json(obj, target=None)[source]¶

Serialize a scikit-chem object as json.