normMethod: Definition of normalization procedure
In qsea: IP-seq data analysis and vizualization

Description Usage Arguments Details Value Author(s) See Also Examples

This function allows to define normalization methods by specifying components.

1	normMethod(methods, ...)

`methods`	names of predefined normalization methods ( for a list of predefined methods, see details)
`...`	sets of normalization components, that can be combined to user defined normalization methods

Predefined normalization methods:

“counts”: no normalization, simply raw count values

“reads”: same as counts

“rpm”: reads per million mappable reads

“nrpm”: CNV normalized reads per million mappable reads

“beta”: transformation to % methylation, posterior mean point estimator

“logitbeta”: logit transformed beta values

“betaLB”: 2.5 lower bound for the point estimator

“betaUB”: 97.5 upper bound for the point estimator

Allowd components for user defined normalization methods:

“library_size”: scale by effective library size

“region_length”: scale by window size

“preference”: scale by positional sequencing preference

“cnv”: scale by CNV ratio

“enrichment”: use enrichment profiles for transformation to absolute methylation level

“qXY”: quantile estimator for transformation to absolute methylation level. XY must be replaced by the quantile (see example with self defined lower and upper bound)

“offset”: consider background reads

WARNING: not all combinations are allowed (eg qXY requires enrichment) and not all allowed combinations are meaningful. Inexperienced users should stick to predefined normalization methods.

a list object, containing the components for the specified normalization procedure

Mathias Lienhard

makeTable

#simply raw counts
nm=normMethod("counts")
#beta-values (% methylation) including lower and upper bounds
nm=normMethod(c("beta", "betaLB", "betaUB"))
#self defined lower and upper bound: 10% and 90% quantile
nm=normMethod("beta", 
betaLB_10=c("enrichment", "cnv", "library_size",
    "region_length", "preference","q10", "offset"),
betaUB_90=c("enrichment", "cnv", "library_size",
    "region_length", "preference","q90", "offset")
)