spliceR (version 1.14.0)

preSpliceRFilter: Filters on spliceR-lists for reduction of data sets

Description

Applies a number of filters on a spliceR object to reduce data set size before running downstream analyses.

Usage

preSpliceRFilter(spliceRobject, filters, expressionCutoff=0)

Arguments

spliceRobject

a SpliceRList object, either created manually from transcript and exon information (see SpliceRList), or created by prepareCuff from CuffLinks data.

filters

vector, giving the filters that should be applied - any combinations of 'geneOK', 'expressedGenes', 'sigGenes', 'isoOK', 'expressedIso', 'isoClass' and/or 'sigIso'. Works only for data from cufflinks, as a manually generated SpliceRList does not include these metacolumns.

expressionCutoff

Numeric, giving the expression threshold (often in FPKM) used for the 'expressedGenes' and 'expressedIso' filter. Default value is 0.

Value

A SpliceRList with transcripts after filtering.

Details

Often, many genes and isoforms are flagged as not "OK" or "LOWDATA" by Cufflinks, indicating low confidence in these. This function is handy for reducing the data size of a Cufflinks data set to reduce running times for downstream analyses.

Note, that preSpliceRFilter removes trancsripts from the dataset permanently, reducing size, while the filter options of spliceR and annotatePTC only selects transcripts for analysis, but does not remove any data.

References

Vitting-Seerup K , Porse BT, Sandelin A, Waage J. (2014) spliceR: an R package for classification of alternative splicing and prediction of coding potential from RNA-seq data. BMC Bioinformatics 15:81.

Examples

#Load cufflinks example data
cuffDB <- prepareCuffExample()

#Generate SpliceRList from cufflinks data
cuffDB_spliceR <- prepareCuff(cuffDB)

#Filter 
cuffDB_spliceR_filtered <- preSpliceRFilter(cuffDB_spliceR, filters=c("expressedIso", "isoOK", "expressedGenes", "geneOK"))