filterDU {ASpli}R Documentation

Filtering ASpliDU objects

Description

ASpliDU object can be filtered to retain genes, bins or junction according to their fdr corrected p-value estimated and log-fold-change.

Usage

 
  filterDU( 
    du , 
    what = c( 'genes','bins','junctions'),
    fdr = 1,
    logFC = 0,
    absLogFC = TRUE,
    logFCgreater = TRUE
  )

Arguments

du

An ASpliDU object

what

A character vector that specifies the kind of features that will be filtered. Accepted values are 'genes', 'bins', 'junctions'. Multiple values can be passed at the same time. The default value is c( 'genes','bins', 'junctions')

fdr

A double value representing the maximum accepted value of fdr corrected p-value to pass the filter. The default value is 1, the neutral value for fdr filtering operation.

logFC

A double value representing the cut-off for accepted values of log-fold-change to pass the filter. The default value is 0, the neutral value for logFC filtering operation if logFCgreater and absLocFC arguments are both TRUE.

absLogFC

A logical value that specifies that the absolute value of log-fold-change will be used in the filter operation. The default value is TRUE.

logFCgreater

A logical value that specifies that the log-fold-change value ( or abs(log-fold-change) if absLogFC argument is TRUE) of features must be greater than the cut-off value to pass the filter. The default value is TRUE.

Value

A new ASpliDU object with the results of the filtering operations. The elements of features that were not specified to be filtered are kept from the input ASpliDU object.

Author(s)

Estefania Mancini, Javier Iserte, Marcelo Yanovsky, Ariel Chernomoretz

See Also

DUreport, junctionDUreport, DUreportBinSplice,

Examples


 # Create a transcript DB from gff/gtf annotation file.
  # Warnings in this examples can be ignored. 
  library(GenomicFeatures)
  genomeTxDb <- makeTxDbFromGFF( system.file('extdata','genes.mini.gtf', 
                                 package="ASpli") )
  
  # Create an ASpliFeatures object from TxDb
  features <- binGenome( genomeTxDb )
  
  # Define bam files, sample names and experimental factors for targets.
  bamFileNames <- c( "A_C_0.bam", "A_C_1.bam", "A_C_2.bam", 
                     "A_D_0.bam", "A_D_1.bam", "A_D_2.bam" )
  targets <- data.frame( 
               row.names = paste0('Sample_',c(1:6)),
               bam = system.file( 'extdata', bamFileNames, package="ASpli" ),
               factor1 = c( 'C','C','C','D','D','D') )
  
  # Load reads from bam files
  bams <- loadBAM( targets )
  
  # Read counts from bam files
  counts   <- readCounts( features, bams, targets, cores = 1, readLength = 100, 
                          maxISize = 50000 )
  
  # Calculate differential usage of junctions only 
  du       <- DUreport( counts, targets )
  
  # Filter by FDR
  duFiltered1 <- filterDU( du, what=c('genes','bins'), 
     fdr = 0.01 )


  # Filter by logFC, only those that were up-regulated 
  duFiltered2 <- filterDU( du, what=c('genes','bins'), 
     logFC = log( 1.5, 2 ),  absLogFC = FALSE )


[Package ASpli version 1.8.1 Index]