pylimma.normalize_between_arrays

pylimma.normalize_between_arrays(object, method=None, targets=None, cyclic_method='fast', *, out_layer='normalized', uns_key='normalize_between_arrays', layer=None, **kwargs)[source]

Normalize columns of an expression matrix between arrays.

Parameters:

  • object (ndarray) – Expression matrix, shape (n_genes, n_samples).

  • method ({"none", "scale", "quantile", "cyclicloess"}, optional) – Normalization method. Defaults to "quantile" for matrix input (matches R’s default for matrices).

  • targets – Ignored for single-channel matrix input. Accepted for R signature parity.

  • cyclic_method (str, default "fast") – Sub-method for cyclic LOESS. Currently only "fast" is implemented.

  • **kwargs – Forwarded to the underlying method (ties for quantile; span, iterations, adaptive_span for cyclic LOESS).

  • out_layer (str)

  • uns_key (str)

  • layer (str | None)

Returns:

Normalized matrix, same shape as input.

Return type:

ndarray

Notes

Two-channel methods ("Aquantile", "Gquantile", "Rquantile", "Tquantile") are not supported; pylimma operates on single-channel expression matrices only.