torchcell.datasets.scerevisiae.oduibhir2014.SmfODuibhir2014Dataset
- class SmfODuibhir2014Dataset(root: str = 'data/torchcell/smf_oduibhir2014', io_workers: int = 0, transform: Callable[[...], Any] | None = None, pre_transform: Callable[[...], Any] | None = None, **kwargs: Any)[source]
Bases:
ExperimentDatasetO’Duibhir 2014 single-deletion relative-growth-rate -> fitness dataset.
- 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
- property experiment_class: type[Experiment]
Experiment schema class produced by this dataset.
- property reference_class: type[ExperimentReference]
Experiment-reference schema class produced by this dataset.
- property raw_file_names: list[str]
The pinned Supplementary Dataset S2 text file (single-mutant relative growth).
- download() None[source]
Copy the pinned Dataset S2 file into raw_dir and verify its sha256.
Publisher-side retrieval (Wiley/EMBO ESM, PMC) is not reliably scriptable, so the canonical source is the sha256-pinned file already staged in the library mirror.
- process() None[source]
Convert each Dataset S2 row into a fitness record; write LMDB.
Every record is built and every refusal raised BEFORE the store is opened, so a refused file leaves no
processed/lmdbbehind for a retry to serve as empty. Every row is a record under its released ORF (upper-cased; no genome is consulted, #1045); a cell that is not ORF-shaped goes on the released-gene-name leaf verbatim (0 of 1312). Refusals (each 0 of the 1312 rows of the pinned file): a blankcommonNameorlog2relTand two rows naming one ORF.