Estimate.EDR.from.pilot: Predict EDR

In MasakiLin/RNASeqDesign: A framework for RNA-Seq genome-wide power calculation and study design issues

Description

Predict EDR based on a pilot data.

Usage

Estimate.EDR.from.pilot(Data, status, group.name = c("Control", "Case"), FDR,
  M, target.N, target.R = NULL, target.theta = NULL, tol = 0.1,
  filter = T, filter.level = 5)

Arguments

Data	Input pilot data.
status	A vector of group labels.
group.name	A vector of length two. First element is the group name of first group, and the second element is of second group.
FDR	FDR level.
M	Number of iterations for parametric bootstrap.
target.N	Targeted sample size.
target.R	Targeted sequencing depth.
target.theta	Targeted proportion of sample size in second group compared to first group.
tol	Threshold for deciding if use CDD to estimate the proportion of DE genes.
filter	Filter genes based on mean counts?
filter.level	Filter the genes with mean counts less than this level.

Value

An object of class list is returned: Result is a list of length M (depends on how many times of parametric bootstrap you run), and each component is also a list of length of variable target.R. Each component contains a matrix with target N in row and summary statistics in columns. Summary statistics provided are TP, TN, EDR.

Author(s)

Chien-Wei Lin

MasakiLin/RNASeqDesign documentation built on May 7, 2019, 3:41 p.m.