torchcell.datasets.scerevisiae.costanzo2016.DmfCostanzo2016Dataset

class DmfCostanzo2016Dataset(root: str = 'data/torchcell/dmf_costanzo2016', subset_n: int | None = None, batch_size: int = 10000, io_workers: int = 1, transform: Callable[[...], Any] | None = None, pre_transform: Callable[[...], Any] | None = None, **kwargs: Any)

Bases: ExperimentDataset

Double-mutant fitness dataset from Costanzo et al. 2016.

organism: ClassVar[Organism] = 'scerevisiae'

each loader declares its own, the build’s organism switch reads it off the class (dataset_organism), and a loader that forgot it is refused by name instead of inheriting an organism it may not have.

Type:

The host organism every record describes. Annotated here WITHOUT a value

download() → None

Download and unpack the Costanzo archive, keeping the SGA text files.

property experiment_class: type[FitnessExperiment]

Return the FitnessExperiment type for DMF data.

property reference_class: type[FitnessExperimentReference]

Return the FitnessExperimentReference type for DMF data.

property raw_file_names: list[str]

Return the SGA interaction text files used as raw input.

preprocess_raw(df: DataFrame, preprocess: dict[str, Any] | None = None) → DataFrame

Concatenate and clean the SGA files into a double-mutant fitness table.

A blank Double mutant fitness or Double mutant fitness standard deviation raises BlankDoubleMutantFitnessError before any record or reference noise is computed.

process() → None

Preprocess the raw SGA files and write records into LMDB in batches.

static create_experiment(dataset_name: str, row: Series, phenotype_reference_std_26: Any, phenotype_reference_std_30: Any) → tuple[FitnessExperiment, FitnessExperimentReference, Publication]

Build DMF experiment, reference, and publication objects for a row.