Reparameterize eSVD object

Usage

reparameterization_esvd_covariates(
  input_obj,
  fit_name,
  omitted_variables = NULL,
  verbose = 0
)

Arguments

input_obj: eSVD object, after either initialize_esvd() or opt_esvd()
fit_name: The name of the fit in input_obj that you wish to reparameterize. This should be in names(input_obj)
omitted_variables: Either NULL (the default) or variables in input_obj$covariates that should not be reparameterized
verbose: Integer.

Value

eSVD object after adjusting the fit in fit_name.

Examples

set.seed(10)
sim <- generate_null(cell_per_person = 15, num_genes = 40,
                     num_individuals = 8)
esvd_obj <- initialize_esvd(dat = sim$obs_mat,
                            covariates = sim$covariates,
                            metadata_individual = sim$metadata_individual,
                            case_control_variable = "CC",
                            bool_intercept = TRUE,
                            k = 2,
                            lambda = 0.1)
esvd_obj <- reparameterization_esvd_covariates(input_obj = esvd_obj,
                                               fit_name = "fit_Init",
                                               omitted_variables = "Log_UMI")
# x_mat is now orthogonal to the retained covariates
max(abs(crossprod(esvd_obj$fit_Init$x_mat,
                  esvd_obj$covariates[, "Sex"])))
#> [1] 1.14353e-13