Module molcrawl.protein_sequence.utils.configs

Classes

class ProteinSequenceConfig (data_preparation: UniProtPreparationConfig = <factory>)

Expand source code

@dataclass
class ProteinSequenceConfig(Config):
    data_preparation: UniProtPreparationConfig = field(default_factory=UniProtPreparationConfig)

    def __post_init__(self):
        if not isinstance(self.data_preparation, UniProtPreparationConfig):  # type: ignore[misc]
            self.data_preparation = UniProtPreparationConfig(**self.data_preparation)  # type: ignore[arg-type]

ProteinSequenceConfig(data_preparation: molcrawl.protein_sequence.utils.configs.UniProtPreparationConfig = )

Ancestors

• Config

Instance variables

var data_preparation : UniProtPreparationConfig

Inherited members

• Config:
  • from_file

class UniProtPreparationConfig (dataset: str = 'UniRef50',
output_dir: str = 'learning_source_dummy/protein_sequence',
use_md5: bool = False,
num_worker: int = 4,
max_lines_per_file: int = 1000000)

Expand source code

@dataclass
class UniProtPreparationConfig:
    # Which uniprot dataset to download must be one of the following:
    # "UniprotKB_reviewed", "UniprotKB_unreviewed", "UniRef100", "UniRef90", "UniRef50", "UniParc"
    dataset: str = "UniRef50"
    # Output directory where the preparation will be made
    output_dir: str = PROTEIN_SEQUENCE_DIR

    # If True use md5 to check if a file needs to be downloaded again, using md5
    # is very time consuming for large file. Otherwise we only check if the path exists.
    use_md5: bool = False

    # Special case for Uniparc download, num of worker to use.
    num_worker: int = 4

    # Number of sequence per files for raw files and parquet. It also reflex the number
    # of sequence loaded in memory during the processing of those files.
    max_lines_per_file: int = 10**6

UniProtPreparationConfig(dataset: str = 'UniRef50', output_dir: str = 'learning_source_dummy/protein_sequence', use_md5: bool = False, num_worker: int = 4, max_lines_per_file: int = 1000000)

Instance variables

var dataset : str

var max_lines_per_file : int

var num_worker : int

var output_dir : str

var use_md5 : bool