checkDataDistribution: fit different count models for a data matrix

In chenlab-sj/nbid: Differential expression analysis using negative binomial with independent dispersions (NBID)

Description

This function fits the count with different count models including Poisson, negative binomial and zero inflated negative binomial models. It invokes checkCountDistributionPerGroup for the real fitting

Usage

checkDataDistribution(data, groups, countPerCell = NULL,
  covariates = NULL, largestKRemoved = 0, ncore = 1)

Arguments

data	the gene-cell data matrix, it assumes the row names of the matrix are the gene names
groups	the group ID vector. If there are multiple groups, the results will be put together with cbind
countPerCell	the total UMI per cell. If it is NULL, it is calculated by summing over all the gene counts for each cell.
covariates	the covariates to account for
largestKRemoved	remove the numbers correspond to the largest K
ncore	the number of cores to use for parallel running

Value

a data frame with fitted information

Examples

## Not run: 
gene1 = rnbinom(100, size = 0.1, mu = 0.1)
gene2 = rpois(100, lambda = 0.1)
data = rbind(gene1, gene2)
groups = rep(1, 100)
countPerCell = rep(1e4, 100)
fitResult = checkDataDistribution(data, groups, countPerCell)
print(t(fitResult))

## End(Not run)

chenlab-sj/nbid documentation built on Nov. 4, 2019, 8:50 a.m.