Skip to contents

PCA on MetaCells (sparse data)

rs_mc_pca.Rd

[Experimental] Calculates PCA for MetaCells or more generally speaking sparse data. This is happening in-memory compared to the (usually much) larger single cell data sets. The matrix is densified, optionally CLR transformed and scaled according to pca_params before the SVD.

Usage

rs_mc_pca(sparse_data, no_pcs, pca_params, clr_offsets, seed, verbose)

Arguments

sparse_data

A named list that needs to have data, indptr, indices, nrow, ncol and cs_type. Shape is (metacells, genes), holding the normalised counts of the genes to use.

no_pcs

Integer. Number of PCs to return.

pca_params

Named list. Contains the parameters to use for this PCA run, see params_sc_pca().

clr_offsets

Optional numeric. One offset per meta cell for the PFlogPF normalisation from Booeshaghi, et al., computed against the full gene panel. Required if pca_params$clr is TRUE, ignored otherwise.

seed

Integer. Random seed for the randomised SVD.

verbose

Integer. 0L - quiet; 1L - normal verbosity; 2L - detailed verbosity.

Value

A list with the following items

  • scores - The samples projected on the PCA space (solved via sparse SVD).

  • loadings - The loadings of the features for the PCA (solved via sparse SVD).

  • singular_values - The singular values for the PCA (solved via sparse SVD).

References

Booeshaghi, et al., bioRxive, 2026.