estNormFactors: Estimate normalization factors

In DSS: Dispersion shrinkage for sequencing data

Description

This function estimates normalization factors for the input 'seqCountSet' object and return the same object with normalizationFactor field filled or replaced.

Usage

## S4 method for signature 'SeqCountSet'
estNormFactors(seqData, method=c("lr", "quantile", "total", "median"))

Arguments

seqData

An object of "SeqCountSet" class.

method

Methods to be used in computing normalization factors. Currently available options only include methods to compute normalization factor to adjust for sequencing depths. Available options use (1) "lr": using median of logratio of counts. Similar to the TMM method. (2) "quantile" (default): 75th quantile, (3) "total": total counts, or (4) "median": median counts to constuct the normalization factors. From all methods the normalization factor will be a vector with same length as number of columns for input counts.

Value

The same "SeqCountSet" object with normalizationFactor field filled or replaced.

Author(s)

Hao Wu <[e-mail]>

Examples

data(seqData)
## compare different methods
seqData=estNormFactors(seqData, "lr")
k1=normalizationFactor(seqData)
seqData=estNormFactors(seqData, "quantile")
k2=normalizationFactor(seqData)
seqData=estNormFactors(seqData, "total")
k3=normalizationFactor(seqData)
cor(cbind(k1,k2,k3))

## assign size factor
normalizationFactor(seqData)=k1

## or normalization factor can be a matrix
dd=exprs(seqData)
f=matrix(runif(length(dd), 1,10), nrow=nrow(dd), ncol=ncol(dd))
normalizationFactor(seqData)=f
head(normalizationFactor(seqData))

DSS documentation built on Nov. 8, 2020, 7:44 p.m.