rxn.reaction_preprocessing.config.RxnImportConfig

class rxn.reaction_preprocessing.config.RxnImportConfig(input_file='${data.path}', output_csv='${data.proc_dir}/${data.name}.imported.csv', data_format=InitialDataFormat.CSV, input_csv_column_name='${common.reaction_column_name}', reaction_column_name='${common.reaction_column_name}', fragment_bond='${common.fragment_bond}', remove_atom_mapping=True, column_for_light=None, column_for_heat=None, keep_original_rxn_column=False)

Bases: object

Configuration for the initial import of the reaction data.

Fields:

input_file: the input file path (.txt, .csv). output_csv: the output file path. initial_data_format: whether the input file is in TXT or CSV format. reaction_column_name: name the column containing the reactions if the input

is in CSV format. The value is ignored if the input is not in CSV format.

reaction_column_name: how to name the column with the reaction SMILES in

the output CSV.

fragment_bond: token denoting a fragment bond in the reaction SMILES (after import). remove_atom_mapping: whether to remove the atom mapping from the input reaction SMILES. column_for_light: name of the column containing boolean values that indicate

whether the reaction happens under light. If specified, the light token will be added to the precursors of the corresponding reactions.

column_for_heat: name of the column containing boolean values that indicate

whether the reaction happens under heat. If specified, the heat token will be added to the precursors of the corresponding reactions.

keep_original_rxn_column: determines whether the original column with the

raw reaction SMILES is to be kept or not.

Parameters

• input_file (str, default: '${data.path}') –

• output_csv (str, default: '${data.proc_dir}/${data.name}.imported.csv') –

• data_format (InitialDataFormat, default: <InitialDataFormat.CSV: 2>) –

• input_csv_column_name (str, default: '${common.reaction_column_name}') –

• reaction_column_name (str, default: '${common.reaction_column_name}') –

• fragment_bond (FragmentBond, default: '${common.fragment_bond}') –

• remove_atom_mapping (bool, default: True) –

• column_for_light (Optional[str], default: None) –

• column_for_heat (Optional[str], default: None) –

• keep_original_rxn_column (bool, default: False) –

__init__(input_file='${data.path}', output_csv='${data.proc_dir}/${data.name}.imported.csv', data_format=InitialDataFormat.CSV, input_csv_column_name='${common.reaction_column_name}', reaction_column_name='${common.reaction_column_name}', fragment_bond='${common.fragment_bond}', remove_atom_mapping=True, column_for_light=None, column_for_heat=None, keep_original_rxn_column=False)

Parameters

• input_file (str, default: '${data.path}') –

• output_csv (str, default: '${data.proc_dir}/${data.name}.imported.csv') –

• data_format (InitialDataFormat, default: <InitialDataFormat.CSV: 2>) –

• input_csv_column_name (str, default: '${common.reaction_column_name}') –

• reaction_column_name (str, default: '${common.reaction_column_name}') –

• fragment_bond (FragmentBond, default: '${common.fragment_bond}') –

• remove_atom_mapping (bool, default: True) –

• column_for_light (Optional[str], default: None) –

• column_for_heat (Optional[str], default: None) –

• keep_original_rxn_column (bool, default: False) –

Return type

None

Methods

__init__([input_file, output_csv, ...])

param input_file

Attributes

column_for_heat
column_for_light
data_format
fragment_bond
input_csv_column_name
input_file
keep_original_rxn_column
output_csv
reaction_column_name
remove_atom_mapping