para_est_new {scpoisson}

Parameter estimates based on two-way approximation

Description

This function returns a vector consists of parameter estimates for overall offset, cell effect, and gene effect.

Usage

para_est_new(test_set)

Arguments

test_set

A UMI count data matrix with genes as rows and cells as columns

Details

This is a function used to calculate parameter estimates based on \lambda_{gc} = e^{\mu + \alpha_g + \beta_c}, where \mu is the overall offset, \alpha is a vector with the same length as the number of genes, and \beta is a vector with the same length as the number of cells. The order of elements in vectors \alpha or \beta is the same as rows (genes) or cells (columns) from input data. Be sure to remove cells/genes with all zeros.

Value

A numeric vector containing parameter estimates from overall offset (first element), gene effect (same order as rows) and cell effect (same order as columns).

Examples

# Matrix as input
test_set <- matrix(rpois(500, 0.5), nrow = 10)
para_est_new(test_set)