Fits a residual model for meta cells — rs_mc_fit

In-memory version of rs_sc_fit_residuals(). Meta cell counts are summed UMIs, so the negative binomial the residual models describe is still defined; it just sits at a much greater depth than a single cell.

Usage

rs_mc_fit_residuals(
  sparse_data,
  method,
  group_of_cell,
  covariates,
  params,
  seed,
  verbose
)

Arguments

sparse_data: A named list that needs to have data, indptr, indices, nrow, ncol and cs_type. Shape is (metacells, genes). Pass raw counts.
method: String. One of c("sctransform", "analytic_pearson").
group_of_cell: Integer vector or NULL. Group label per meta cell. (0-indexed, dense!) NULL fits one model.
covariates: Named list of numeric vectors, one per covariate, each of length nrow. scTransform only.
params: Named list. See params_sc_sctransform() or params_sc_apr().
seed: Integer. Seed for the step-1 subsample.
verbose: Integer. 0L - quiet; 1L - normal verbosity; 2L - detailed verbosity.

Value

A list as described in rs_sc_fit_residuals().