Estimate nuisance values for matrix or sparse matrices. — estimate

Each gene's value is the maximum-likelihood Gamma rate \(\beta_j = 1/\gamma_j\) under \(A_{ji} \mid \lambda_{ji} \sim \mathrm{Poisson}(\ell_{ji} \lambda_{ji})\), \(\lambda_{ji} \sim \mathrm{Gamma}(\mu_{ji} \beta_j, \beta_j)\), with \(\mu\) from mean_mat and \(\ell\) from library_mat. Larger values mean less over-dispersion. The estimate is then bounded above by cap_multiplier times the median of the gene's column of library_mat; see estimate_nuisance.eSVD for why.

Usage

# Default S3 method
estimate_nuisance(
  input_obj,
  mean_mat,
  library_mat,
  bool_use_log = FALSE,
  cap_multiplier = 10,
  min_val = 1e-04,
  verbose = 0,
  ...
)

Arguments

input_obj: Dataset (either matrix or dgCMatrix) where the \(n\) rows represent cells and \(p\) columns represent genes. The rows and columns of the matrix should be named.
mean_mat: A matrix of \(n\) rows and \(p\) columns that represents the expected value of each entry.
library_mat: A matrix of \(n\) rows and \(p\) columns that represents the library size of each entry.
bool_use_log: Boolean if the nuisance (i.e., over-dispersion) parameter should be estimated on the log scale, default is FALSE.
cap_multiplier: One positive number, default 10: no gene's rate may exceed this many times the median of its column of library_mat. Inf means no cap.
min_val: Minimum value of the rate, as a multiple of the median over cells of the gene's library size (the units of cap_multiplier); must be below cap_multiplier. Default 1e-4.
verbose: Integer.
...: Additional parameters.

Value

Numeric vector of length \(p\) (named by colnames(input_obj) when present) of Gamma rates, after the cap. A gene whose estimation fails on both routes gets the floor, min_val times the median of its library size, and a warning reports how many genes did. The rates before the cap and each gene's status are returned by the eSVD method only.