run_model_comparison: Fit regression models for many genes

In churchill-lab/scRATE: A Bayesian model selection approach to test zero inflation from scRNA-seq data

Description

Fit regression models for many genes

Usage

run_model_comparison(cntfile, formula_string = NULL, nCores = NULL,
  seed = NULL, adapt_delta = 0.8, model2fit = NULL, outfile = NULL)

Arguments

cntfile	Expression quantity file (RData format: Use 'save' and 'load'). The count matrix is gene x cell.
formula_string	A regression formula to fit the non-zero-inflated counts
nCores	Number of cores
seed	Seed number
adapt_delta	The target average proposal acceptance probability during Stan’s adaptation period (default:0.8)
model2fit	A specific model to fit (1:P, 2:NB, 3:ZIP, 4:ZINB, NULL:All)
outfile	Output file name to store ELPD_loo results (RDS format)

Value

A list of ELPD_loo results and mean parameters returned by loo::compare

churchill-lab/scRATE documentation built on Aug. 12, 2020, 2:52 p.m.