model_drop_sd: Model Dropout SD as a Quadratic Function of the Mean Dropout

In kdzimm/hierarchicell: Simulating Cell-Type Specific and Hierarchical Single-Cell RNA-Seq Data

Description

This function models gene dropout SD as a quadratic function of the dropout mean. Visualizing this fit and looking for oddities and extreme outliers may be helpful.

Usage

model_drop_sd(data_summaries, plot = FALSE)

Arguments

data_summaries	an R object that has been output by the package's compute_data_summaries function.
plot	a TRUE/FALSE statement for the output of a plot to observe how well gene dropout SD behaves as a quadratic function of mean dropout

Details

Prior to estimating the simulation parameters, it is important to run the compute_data_summaries function to build an object that is in the right format for the following estimation functions to properly compute. The 'ggplot2' package is required for the plotting component of this function to work. Please install it.

Value

A plot (if plot=TRUE) and a vector of length three, where each number is the estimate of the intercept or slope for the model.

Examples

clean_expr_data <- filter_counts()
data_summaries <- compute_data_summaries(clean_expr_data)
gene_drop_betas <- model_drop_sd(data_summaries)

kdzimm/hierarchicell documentation built on Dec. 21, 2021, 5:23 a.m.