LARRY Dataset#

Lineage tracing data from Weinreb et al. (Science, 2020), in which barcoded hematopoietic progenitors are profiled across three timepoints, allowing observed clonal fate to be compared against predicted fate.

import scdiffeq as sdq

adata = sdq.datasets.larry()

See Datasets and data sources for where files are cached and what to cite.

Variants#

Two variants are available, and the difference between them matters:

variant

Shape

X_pca

Description

None (default)

130,887 × 2,492

included

The biology-rich object. Already gene-filtered, and ships precomputed X_pca, X_umap and X_scaled.

"unprocessed"

130,887 × 25,289

not included

The upstream input, before gene filtering. Carries a var["use_genes"] column that preprocessing filters by, reducing it to 2,447 genes. X_pca is computed locally.

 
# default: ships its own X_pca
adata = sdq.datasets.larry()

# the unprocessed input; X_pca is computed during preprocessing
adata = sdq.datasets.larry(variant="unprocessed")

Deprecated since version 1.1.1: variant="fate_prediction" is deprecated in favour of variant="unprocessed". The old name still works and emits a DeprecationWarning.

The rename corrects a misleading name. That variant’s upstream filename is adata.Weinreb2020.in_vitro.gene_filtered.h5ad, but the object it contains is the unfiltered one – the gene_filtered name refers to the presence of the filtering metadata, not to filtering having been applied. Reading it as “already gene-filtered” led to the reasonable but incorrect expectation that it would carry a precomputed X_pca.

Preprocessing#

With default arguments, larry() annotates the object with precomputed CytoTRACE values, filters genes on var["use_genes"], then scales and runs a 50-component PCA:

adata = sdq.datasets.larry(
    filter_genes=True,       # subset to var["use_genes"]
    reduce_dimensions=True,  # StandardScaler + PCA(50, random_state=0)
    cytotrace=True,          # merge precomputed CytoTRACE annotations
)

Each flag is independent; cytotrace=True is honoured even when the other two are disabled. Non-default combinations are cached separately, so reduce_dimensions=False does not overwrite the default cache.

The CytoTRACE annotations are keyed to the gene-filtered gene set, so on the "unprocessed" variant they cover 2,492 of its 25,289 genes; the remainder are NaN. This is expected, not an error.

API#

scdiffeq.datasets._larry_in_vitro.larry(data_dir: str = '/home/runner/work/scDiffEq/scDiffEq/docs/source', variant: str | None = None, filter_genes: bool = True, reduce_dimensions: bool = True, cytotrace: bool = True, force_download: bool = False, force_preprocess: bool = False) → AnnData[source]#

LARRY in vitro dataset

The raw download and the preprocessed result are cached as separate files, so a dataset obtained by any means (including a manual download) is still preprocessed on first use.

Parameters:

  • data_dir – str, default=os.getcwd() Path to the directory where the data will be saved.

  • variant –

    Optional[str], default=None Dataset variant.

    None (default) is the biology-rich object: 130,887 x 2,492, already gene-filtered, shipping a precomputed X_pca, X_umap and X_scaled.

    "unprocessed" is the upstream input: 130,887 x 25,289 with no X_pca, carrying a use_genes column that preprocessing filters by (down to 2,447 genes). X_pca is computed locally, or downloaded prebuilt when available.

    variant="fate_prediction" is a deprecated alias for "unprocessed" and will be removed in a future release. Despite its upstream filename (...in_vitro.gene_filtered.h5ad), that object is the unfiltered one.

  • filter_genes – bool, default=True Whether to subset to adata.var['use_genes']. A no-op for variants that are already gene-filtered.

  • reduce_dimensions – bool, default=True Whether to scale and run PCA (50 components, random_state=0).

  • cytotrace – bool, default=True Whether to annotate with precomputed CytoTRACE values. Applied independently of the other preprocessing flags.

  • force_download – bool, default=False Re-download the raw file and regenerate the processed file.

  • force_preprocess – bool, default=False Regenerate the processed file from the cached raw file without re-downloading it.

Returns:

Preprocessed AnnData object.

Return type:

AnnData