REMProduct-class: REMProduct instances
In YinanZheng/REMP: Repetitive Element Methylation Prediction

REMProduct-class

R Documentation

REMProduct instances

Description

Class REMProduct is to maintain RE methylation prediction results. REMProduct inherits Bioconductor's RangedSummarizedExperiment class.

Usage

REMProduct(
  REtype = "Unknown",
  genome = "Unknown",
  platform = "Unknown",
  win = "Unknown",
  predictModel = "Unknown",
  QCModel = "Unknown",
  rempM = NULL,
  rempB = NULL,
  rempQC = NULL,
  cpgRanges = GRanges(),
  sampleInfo = DataFrame(),
  REannotation = GRanges(),
  RECpG = GRanges(),
  regionCode = DataFrame(),
  refGene = GRanges(),
  varImp = DataFrame(),
  REStats = DataFrame(),
  GeneStats = DataFrame(),
  Seed = NULL
)

rempM(object)

rempB(object)

rempQC(object)

rempAnnot(object)

rempImp(object)

rempStats(object)

remplot(object, ...)

details(object)

decodeAnnot(object, ...)

rempTrim(object, ...)

rempAggregate(object, ...)

rempCombine(object1, object2)

## S4 method for signature 'REMProduct'
rempM(object)

## S4 method for signature 'REMProduct'
rempB(object)

## S4 method for signature 'REMProduct'
rempQC(object)

## S4 method for signature 'REMProduct'
rempImp(object)

## S4 method for signature 'REMProduct'
rempAnnot(object)

## S4 method for signature 'REMProduct'
rempStats(object)

## S4 method for signature 'REMProduct'
remplot(object, type = c("individual", "overall"), ...)

## S4 method for signature 'REMProduct'
details(object)

## S4 method for signature 'REMProduct'
decodeAnnot(object, type = c("symbol", "entrez"), ncore = 1, BPPARAM = NULL)

## S4 method for signature 'REMProduct'
rempTrim(object, threshold = 1.7, missingRate = 0.2)

## S4 method for signature 'REMProduct'
rempAggregate(object, NCpG = 2, ncore = 1, BPPARAM = NULL)

## S4 method for signature 'REMProduct,REMProduct'
rempCombine(object1, object2)

Arguments

`REtype`	Type of RE (`"Alu"`, `"L1"`, or `"ERV"`).
`genome`	Specify the build of human genome. Can be either `"hg19"` or `"hg38"`.
`platform`	Illumina methylation profiling platform (`"450k"` or `"EPIC"`).
`win`	Flanking window size of the predicting RE-CpG.
`predictModel`	Name of the model used for prediction.
`QCModel`	Name of the model used for prediction quality evaluation.
`rempM`	Predicted methylation level in M value.
`rempB`	Predicted methylation level in beta value (optional).
`rempQC`	Prediction quality scores, which is available only when Random Forest model is used in `remp`.
`cpgRanges`	Genomic ranges of the predicting RE-CpG.
`sampleInfo`	Sample information.
`REannotation`	Annotation data for the predicting RE.
`RECpG`	Annotation data for the RE-CpG profiled by Illumina platform.
`regionCode`	Internal index code defined in `refGene` for gene region indicators.
`refGene`	refSeq gene annotation data, which can be obtained by `fetchRefSeqGene`.
`varImp`	Importance of the predictors.
`REStats`	RE coverage statistics, which is internally generated in `remp`.
`GeneStats`	Gene coverage statistics, which is internally generated in `remp`.
`Seed`	Random seed for Random Forest model for reproducible prediction results.
`object`	A `REMProduct` object.
`...`	For `plot`: `graphical parameters` to be passed to the `plot` method.
`object1`	A `REMProduct` object.
`object2`	A `REMProduct` object.
`type`	For `plot` and `decodeAnnot`: see Utilities.
`ncore`	For `decodeAnnot` and `rempAggregate`: number of cores used for parallel computing. By default no parallel computing is allowed (`ncore = 1`).
`BPPARAM`	For `decodeAnnot` and `rempAggregate`: an optional `BiocParallelParam` instance determining the parallel back-end to be used during evaluation. If not specified, default back-end in the machine will be used.
`threshold`	For `rempTrim`: see Utilities.
`missingRate`	For `rempTrim`: see Utilities.
`NCpG`	For `rempAggregate`: see Utilities.

Value

An object of class REMProduct for the constructor.

Accessors

rempM(object): Return M value of the prediction.
rempB(object): Return beta value of the prediction.
rempQC(object): Return prediction quality scores.
rempImp(object): Return relative importance of predictors.
rempStats(object): Return RE and gene coverage statistics.
rempAnnot(object): Return annotation data for the predicted RE.

Utilities

remplot(object, type = c("individual", "overall"), ...): Make a density plot of predicted methylation (beta values) in the REMProduct object. If type = "individual", density curves will be plotted for each of the samples; If type = "overall", one density curve of the mean methylation level across the samples will be plotted. Default type = "individual".
details(object): Display detailed descriptive statistics of the predicion results.
decodeAnnot(object, type = c("symbol", "entrez")), ncore = NULL, BPPARAM = NULL: Decode the RE annotation data by Gene Symbol (when type = "Symbol") or Entrez Gene (when type = "Entrez").Default type = "Symbol". Annotation data are provided by org.Hs.eg.db.
rempTrim(object, threshold = 1.7, missingRate = 0.2): Any predicted CpG values with quality score < threshold (default = 1.7, specified by threshold = 1.7) will be replaced with NA. CpGs contain more than missingRate * 100 rate across samples will be discarded. Relavant summary statistics will be re-evaluated.
rempAggregate(object, NCpG = 2, ncore = NULL, BPPARAM = NULL): Aggregate the predicted RE-CpG methylation by RE using mean. To ensure the reliability of the aggregation, by default only RE with at least 2 predicted CpG sites (specified by NCpG = 2) will be aggregated.
rempCombine(object1, object2): Combine two REMProduct objects by column.