Package 'funtooNorm'

Title: Normalization Procedure for Infinium HumanMethylation450 BeadChip Kit
Description: Provides a function to normalize Illumina Infinium Human Methylation 450 BeadChip (Illumina 450K), correcting for tissue and/or cell type.
Authors: Celia Greenwood <[email protected]>,Stepan Grinek <[email protected]>, Maxime Turgeon <[email protected]>, Kathleen Klein <[email protected]>
Maintainer: Kathleen Klein <[email protected]>
License: GPL-3
Version: 1.29.0
Built: 2024-09-28 05:16:20 UTC
Source: https://github.com/bioc/funtooNorm

Help Index

funtooNorm

Description

The funtooNorm Package provides a normalization method for data arising from the Illumina Infinium Human Methylation 450 BeadChip (Illumina 450K), including explicit considerations of differences between tissues or cell types. This method should only be used when the data set contains samples fom multiple different tissues or cell types.

Details

Package: funtooNorm
Type: Package
License: GPL-3

Author(s)

Celia Greenwood, Stepan Grinek, Raphael Poujol, Maxime Turgeon, Kathleen Oros Klein

Function to measure intra-replicate agreement for methylation data.

Description

Function to measure intra-replicate agreement for methylation data.

Usage

agreement(Beta, individualID)

Arguments

Beta

: Matrix with beta-values, rows corresponding to probes, columns corresponding to samples.
individualID

: a vector where 2 replicates have the exact same value for two technical replicates. Order of samples should nmatch the samples (columns) in Beta

Details

We expect that the values returned by the agreement function after normalization by funtooNorm to be smaller than before.

Value

The average value of the square distance between replicates: a measure of agreement between replicates in methylation data.

Examples

agreement(cbind(rnorm(n = 10),rnorm(n = 10),rnorm(n = 10)),c(1,1,1))

Creates a S4 object of class 'SampleSet' from GenomeStudio files

Description

Creates a S4 object of class 'SampleSet' from GenomeStudio files

Usage

fromGenStudFiles(controlProbeFile, signalFile, cell_type)

Arguments

controlProbeFile

The control probe file exported from GenomeStudio
signalFile

The signals exported from GenomeStudio samples must be in same order as the control probe File
cell_type

A vector of cell types, names must match control probes and signal files.

Value

An object of class 'SampleSet'.

Creates an object of class SampleSet from a RGChannelSet minfi

Description

Creates a object of class SampleSet from the raw unprocessed data in RGChannelSet

Usage

fromRGChannelSet(myRGChannelSet)

Arguments

myRGChannelSet

: RGChannelSet, from minfi package, should contain a cell_type vector in pData

Value

An object of class 'SampleSet'

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)

The funtooNorm normalization function

Description

funtooNorm Returns the normalized signals to the SampleSet object

Usage

funtooNorm(object, type.fits = "PCR", ncmp = 4, force = FALSE,
  sex = NULL)

## S4 method for signature 'SampleSet'
funtooNorm(object, type.fits = "PCR", ncmp = 4,
  force = FALSE, sex = NULL)

Arguments

object

Object of class SampleSet
type.fits

Choice between "PCR" or "PLS" (default="PCR")
ncmp

Number of components included in the analysis (default=4)
force

If set to TRUE, forces the normalization procedure to re-compute
sex

Boolean vector if male. if NULL Beta values from ChrY are used for classification.

Details

This is a generic function which applies to autosomes and the X chromosome. Chromosome Y requires separate analysis as there are few probes on Y. We use a straightforward quantile normalization applied to males only.

Value

a S4 object of class SampleSet containing the normalized signal

Methods (by class)

  • SampleSet: The funtooNorm normalization function

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)
mySampleSet=funtooNorm(mySampleSet)

Build GRange object of methylation probes

Description

Build GRange object of methylation probes

Usage

getGRanges(object)

## S4 method for signature 'SampleSet'
getGRanges(object)

Arguments

object

Object of class SampleSet.

Value

A GRange object of the positions of each cpg.

Methods (by class)

  • SampleSet: Build GRange object of methylation probes

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)
gr=getGRanges(mySampleSet)

Computes Beta values from normalized signals

Description

Computes Beta values from normalized signals

Usage

getNormBeta(object, offset = 100)

## S4 method for signature 'SampleSet'
getNormBeta(object, offset = 100)

Arguments

object

of type SampleSet
offset

default is 100 as Illumina standard

Value

a matrix containing beta after normalization value for each CpG position and each samples

Methods (by class)

  • SampleSet: Computes Beta values from normalized signals

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)
b=getNormBeta(funtooNorm(mySampleSet))

Computes M values,log2(Meth/Unmeth), from normalized signals

Description

Computes M values,log2(Meth/Unmeth), from normalized signals

Usage

getNormM(object)

## S4 method for signature 'SampleSet'
getNormM(object)

Arguments

object

An object of class SampleSet

Value

a matrix containing M values, log2(Meth/Unmeth), after normalization

Methods (by class)

  • SampleSet: Computes M values, log2(Meth/Unmeth), from normalized signals

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)
m=getNormM(funtooNorm(mySampleSet))

Computes Beta value from raw signals

Description

Computes Beta value from raw signals

Usage

getRawBeta(object, offset = 100)

## S4 method for signature 'SampleSet'
getRawBeta(object, offset = 100)

Arguments

object

object of class SampleSet
offset

default is 100 as Illumina standard

Value

a matrix containing the raw beta value for each position and each samples

Methods (by class)

  • SampleSet: Computes Beta value from raw signals

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)
r=getRawBeta(mySampleSet)

Computes M values after normalization of SNP data.

Description

Computes M values after normalization of SNP data.

Usage

getSnpM(object)

## S4 method for signature 'SampleSet'
getSnpM(object)

Arguments

object

of class SampleSet

Value

a matrix containing M values, log2(Meth/Unmeth), after normalization for SNP data

Methods (by class)

  • SampleSet: Computes M values, log2(Meth/Unmeth), for normalized SNP data

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)
snp=getSnpM(funtooNorm(mySampleSet))

plot of Validation Graph for determing number of components

Description

Plots a series of graphs for each signal type, to determine the number of components to include in the normalization procedure.

Usage

plotValidationGraph(object, type.fits = "PCR", pdf.file = NULL)

## S4 method for signature 'SampleSet'
plotValidationGraph(object, type.fits = "PCR",
  pdf.file = NULL)

Arguments

object

of class SampleSet
type.fits

can be "PCR" or "PLS" (default "PCR")
pdf.file

if no file name is provided print pdf file plotValidationGraph.pdf in working directory.

Value

No value is returned. The function prints the plots to a pdf file.

Methods (by class)

  • SampleSet: Plots a series of graphs for each signal type, to determine the number of components to include in the normalization procedure.

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)
plotValidationGraph(mySampleSet)

S4 class object SampleSet

Description

SampleSet is an S4 class defined for the purpose of running the funtooNorm algorithm. They are lists containing signal data and different variables useful for funtooNorm. The data is separated into the 3 probes types, each having 2 channels (methylated and unmethylated ie : A and B) We then define then the 6 (2*3) labels: AIGrn BIGrn AIRed BIRed AII BII

Value

a S4 object of class SampleSet

Slots

type

Character: is 'minfi' or 'GenomeStudio'

sampleNames

character vector: contain the list of sample names in order used

sampleSize

numeric: the number of samples

nPos

numeric: the number of positions in the ILLUMINA chip

annotation

character: the annotation object from minfi package

cell_type

factor: vector of the cell type for each sample as factors

qntllist

numeric: vector of ordered quantiles

quantiles

list: list of 6 quantiles tables for the 6 signal types

ctl.covmat

matrix: covariance matrix for the model fit

signal

list: list of the values for all 6 probe types.

names

list: list of probes for each type

predmat

list: list of the normalized values for all 6 probe types.

Examples

showClass("SampleSet")

Show Object SampleSet

Description

Display informations about the SampleSet object

Usage

## S4 method for signature 'SampleSet'
show(object)

Arguments

object

an object of class SampleSet
...

optional arguments passed to or from other methods.

Value

No value is returned. The function prints the summary of object of class SampleSet to screen

Examples

require(minfiData)
pData(RGsetEx)$cell_type <- rep(c("type1","type2"),3)
mySampleSet=fromRGChannelSet(RGsetEx)
mySampleSet