bigmelon-normalization: Bigmelon Quantile Normalization methods.

In bigmelon: Illumina methylation array analysis for large experiments

Description

Set of functions that are used to perform quantile normalization methods on gds.class objects.

Usage

## S4 method for signature 'gds.class'
dasen(mns, fudge = 100, ret2 = FALSE, node="betas",...)
dasen.gds(gds, node, mns, uns, onetwo, roco, fudge, ret2)
qn.gdsn(gds, target, newnode)
db.gdsn(gds, mns, uns)
dfsfit.gdsn(gds, targetnode, newnode, roco, onetwo)

Arguments

gds	A gds.class object
node	The "name" of desired output gdsn.class node
mns	The gdsn.class node that corresponds to "methylated" intensities.
uns	The gdsn.class node that corresponds to "unmethylated" intensities.
onetwo	The gdsn.class node that corresponds to probe designs (in reference to 450k and EPIC arrays) OR character string pointing to location of gdsn.class node. e.g. "fData/DESIGN" OR vector containing probe design types of length > 1.
roco	This allows a background gradient model to be fit. This is split from data column names by default. roco=NULL disables model fitting (and speeds up processing), otherwise roco can be supplied as a character vector of strings like 'R01C01' (only 3rd and 6th characters used).
fudge	The value added to total intensity to prevent denominactors close to zero when calculation betas. default = 100
ret2	if TRUE, appends the newly calculated methylated and unmethylated intensities to original gds (as specified in gds arguement). Will overwrite the raw intensities.
target	Target gdsn.class node to perform normalization on. If using "*****.gds" method you do not need to specify this.
targetnode	Target gdsn.class node to perform normalization on. If using "*****.gds" method you do not need to specify this.
newnode	"name" of desired output gdsn.class node. If using "*****.gds" method you do not need to specify this.
...	Additional args such as roco or onetwo.

Details

Each function performs a normalization method described within the wateRmelon package. Functions: qn.gdsn, design.qn.gdsn, db.gdsn and dfsfit.gdsn are described to allow users to create their own custom normalization methods. Otherwise calling dasen or dasen.gds e.t.c will perform the necessary operations for quantile normalization.

Each 'named' normalization method will write a temporary gds object called "temp.gds" into the current working directory and it is removed when normalization is complete. Current methods supplied by default arguments will replace the raw betas with normalized betas, but leave the methylated and unmethylated intensities unprocessed.

Value

Normalization methods return nothing but will affect the gds file and replace/add nodes given to the function.

Author(s)

Tyler J Gorrie-Stone <[e-mail]>

See Also

wateRmelon, dasen

Examples

data(melon)
e <- es2gds(melon,'wat_melon.gds')
dasen(e)
closefn.gds(e) # Close gds object
unlink('wat_melon.gds') # Delete Temp file

bigmelon documentation built on Nov. 8, 2020, 7:40 p.m.