torchcell.datasets.scerevisiae.messner2023.ProteomeMessner2023Dataset

class ProteomeMessner2023Dataset(root: str = 'data/torchcell/proteome_messner2023', io_workers: int = 0, transform: Any | None = None, pre_transform: Any | None = None, **kwargs: Any)

Bases: ExperimentDataset

Microflow-SWATH-MS proteome of the yeast gene-KO collection (4,699 strains).

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 processed matrix + sample metadata required before processing.

download() → None

Copy the hash-pinned matrix + metadata from the library mirror to raw_dir.

The mirror is the canonical source (Mendeley fetched once); each file’s sha256 is verified before the copy (and again at build time by process), and no live Mendeley URL is touched.

process() → None

Aggregate the proteome matrix into per-strain experiments and write LMDB.

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

Preprocessing is handled inside process() for this dataset.

create_experiment(row: dict[str, Any]) → tuple[ProteinAbundanceExperiment, ProteinAbundanceExperimentReference, Publication]

Build the ProteinAbundance experiment/reference/publication for one strain.