Learn R Programming

rnaseqcomp (version 1.2.2)

signalCalibrate: Quantification Filtering And Calibration

Description

This is the function to do any pre-filtering or pre-processing analysis for downstream benchmark estimation and visualization. Pre-filtering includes row selection (e.g. protein coding genes) of quantification table; pre-processing includes calculation on a set of rows as calibration reference (e.g. house keeping genes) across different quantification pipelines, calibration of quantifications across all the pipelines based on given cutoffs from selected pipelines.

Usage

signalCalibrate(quantData, condInfo, repInfo, evaluationFeature,
  calibrationFeature, unitReference, unitCutoff = 0,
  calibrationFeature2 = NULL, fixMedian = 4.776)

Arguments

quantData
A list of quantification matrices each with rows by features (transcripts, genes, junctions or exons) and columns by samples. Names of the list should be provided. The sizes of each element should be the same. Missing data will be set to 0.
condInfo
A factor documenting condition information of samples, corresponding to the columns of each element in quantData.
repInfo
A factor documenting replicate information of samples, corresponding to the columns of each element in quantData.
evaluationFeature
A logical vector corresponding to the rows of each element in quantData, providing which features should be considered for downstream evaluation, e.g. protein coding genes.
calibrationFeature
A logical vector corresponding to the rows of each element in quantData, providing which features should be considered as calibration reference, e.g. house keeping genes.
unitReference
A numeric number specifying which pipeline will be selected as reference pipeline, i.e. the index of one element in quantData.
unitCutoff
A numeric number for signal cutoff on reference pipeline specified by unitReference (default: 0). Equivalent effects of cutoffs will be applied to other pipelines accordingly.
calibrationFeature2
A logical vector corresponding to the rows of each element in quantData, providing which features should be considered as references for calibration across different datasets. Default NULL means no calibration needed.
fixMedian
A numeric number specifying the median of detrend logsignals for features specified by calibrationFeature2. When comparing across datasets, those features will be calibrated to have the same median as fixMedian, while other features calibrated accordingly. The default is 4.776, which was calculated based on one ENCODE dataset used in our web tool.

Value

  • A rnaseqcomp S4 class object
  • quantDataA filtered and calibrated list of quantifications for downstream analysis.
  • condInfoA factor documenting sample condition information.
  • repInfoA factor documenting sample replicate information.
  • refMedA list of numeric vectors giving the log scale medians of calibration features in different pipelines.
  • scalerA number that was used for scaling quantifications onto reference pipeline.

Details

In the functions plotSD and plot2TX, detrended signals with value 0 will be at the same level as value 1 for giving pipeline by unitReference.

Examples

Run this code
data(simdata)
condInfo <- factor(simdata$samp$condition)
repInfo <- factor(simdata$samp$replicate)
evaluationFeature <- rep(TRUE, nrow(simdata$meta))
calibrationFeature <- simdata$meta$house & simdata$meta$chr == 'chr1'
unitReference <- 1
dat <- signalCalibrate(simdata$quant, condInfo, repInfo, evaluationFeature,
calibrationFeature, unitReference, calibrationFeature2 = calibrationFeature)

Run the code above in your browser using DataLab