Initialize Data-driven Prior Covariance Matrices using PCA — pois_cov

Initialize data-driven prior covariance matrices based on principal component analysis.

pois_cov_init(
  data,
  ruv = FALSE,
  Fuv = NULL,
  rho = NULL,
  prop = 1,
  seed = 1,
  npc = 5,
  cutoff = 3
)

Arguments

data: “pois.mash” data object, typically created by calling pois_mash_set_data.
ruv: Logical scalar indicating whether to account for unwanted variation. Default is FALSE. If ruv = TRUE, Fuv and rho must be provided.
Fuv: J x D matrix of latent factors causing unwanted variation, with features as rows and latent factors as columns.
rho: D x R matrix of effects corresponding to unwanted variation, such that bias = Fuv %*% rho.
prop: The proportion by which to take a random subset of genes for prior covariance estimation (useful in case of many genes).
seed: Useful for reproducibility when prop is less than 1.
npc: The number of principal components to use.
cutoff: The threshold for the maximum of absolute values of Z-scores taken across conditions to include as "strong" features used for prior covariance estimation.

A list with initial estimates of prior covariances, and indices of the features (j = 1,...,J) to include in the subsequent ED step.