Estimate abundances cond

Predict cell type abundances given a PLN model and a set of inputs for its covariates and observed counts

estimate_abundances_cond(
  ccm,
  newdata,
  cond_responses,
  min_log_abund = -5,
  cell_group = "cell_group",
  type = c("link", "response"),
  pln_model = c("full", "reduced")
)

Arguments

ccm

A cell_count_model, fit with covariance_type = "full".

newdata

tibble A tibble of variables used for the prediction. Must either be a single row or a tibble with one row per sample of the cell count set for ccm.

cond_responses

A data frame containing the observed counts of the conditioning cell type(s); column names must match cell types in the model.

min_log_abund

numeric Minimum log abundance value.

cell_group

string The name of the groups that are being estimated.

type

string Scale of the predictions. Either "link" (default, latent scale) or "response" (mean response scale).

pln_model

string Which PLN model form to use for prediction. Either "full" (default) or "reduced".

Value

A tibble of cell abundance predictions for the unconditioned cell types.