runPathwaySplice: runPathwaySplice

In SCCC-BBC/PathwaySplice: An R Package for Unbiased Splicing Pathway Analysis in RNAseq data

Description

This function identifies pathways that are enriched with signficant genes, while accounting for different number of gene features (e.g. exons) associated with each gene

Usage

runPathwaySplice(genewise.table, genome, id, gene2cat = NULL,
  test.cats = c("GO:CC", "GO:BP", "GO:MF"), go.size.limit = c(10, 200),
  method = "Wallenius", repcnt = 2000, use.genes.without.cat = FALSE,
  binsize = "auto", output.file = tempfile())

Arguments

genewise.table	data frame returned from function makeGeneTable
genome	Genome to be used, options are 'hg19' or 'mm10'
id	GeneID, options are 'entrezgene' or 'ensembl_gene_id'
gene2cat	Get sets to be tested, these are defined by users, can be obtained from gmtGene2Cat function
test.cats	Default gene ontology gene sets to be tested if gene2cat is not defined
go.size.limit	Size limit of the gene sets to be tested
method	the method used to calculate pathway enrichment p value. Options are 'Wallenius', 'Sampling', and 'Hypergeometric'
repcnt	Number of random samples to be calculated when 'Sampling' is used, this argument ignored unless method='Sampling'
use.genes.without.cat	Whether genes not mapped to any gene_set tested are included in the analysis. Default is set to FALSE, where genes not mapped to any tested categories are ignored in analysis. Set this option to TRUE if it's desired that all genes in genewise.table to be counted towards the total number of genes outside the category.
binsize	The number of genes in each gene bin in the bias plot
output.file	File name for the analysis result in .csv format.

Details

This function implements the methodology described in Young et al. (2011) to adjust for different number of gene features (column numFeature in gene.based.table). For example, gene features can be non-overlapping exon counting bins associated with each gene (Fig 1 in Anders et al. 2012). In the bias plot, the genes are grouped by numFeature in genewise.table into gene bins, the proportions of signficant genes are then plotted against the gene bins.

Value

runPathwaySplice returns a tibble with the following information:

gene_set	Name of the gene set. Note in this document we used the terms gene_set, category, and pathway interchangeably
over_represented_pvalue	P-vaue for the associated gene_set being over-represented among significant genes
under_represented_pvalue	P-vaue for the associated gene_set being under-represented among significant genes
numSIGInCat	The number of significant genes in the gene_set
numInCat	The total number of genes in the gene_set
description	Description of the gene gene_set
ontology	The domain of the gene ontology terms if GO categories were tested. Go categories can be classified into three domains: cellular component, biological process, molecular function.
SIGgene_ensembl	Ensembl gene ID of significant genes in the gene_set
SIGgene_symbol	Gene symbols of signficant genes in the gene_set
Ave_value_all_gene	The average value for numFeature for all the genes in the gene_set, note that numFeature is the bias factor adjusted by PathwaySplice

These information are also saved in the file output.file

References

Young MD, Wakefield MJ, Smyth GK, Oshlack A (2011) Gene ontology analysis for RNA-seq: accounting for selection bias. Genome Biology 11:R14

Anders S, Reyes A, Huber W (2012) Dececting differential usage of exons from RNA-seq data. Genome Research 22(10): 2008-2017

Examples

gene.based.table <- makeGeneTable(featureBasedData)

res <- runPathwaySplice(gene.based.table,genome='hg19',id='ensGene',
                         test.cats=c('GO:BP'),
                         go.size.limit=c(5,30),
                         method='Wallenius',binsize=20)
## Not run: 

# demonstrate how output file can be specified                 
res <- runPathwaySplice(gene.based.table,genome='hg19',id='ensGene',
                       test.cats=c('GO:BP'),
                       go.size.limit=c(5,30),
                       method='Wallenius',binsize=800, 
                       output.file=tempfile())    

# demonstrate using customized gene sets
dir.name <- system.file('extdata', package='PathwaySplice')
hallmark.local.pathways <- file.path(dir.name,'h.all.v6.0.symbols.gmt.txt')
hlp <- gmtGene2Cat(hallmark.local.pathways, genomeID='hg19')

res <- runPathwaySplice(gene.based.table,genome='hg19',id='ensGene',
                       gene2cat=hlp,
                       go.size.limit=c(5,200),
                       method='Wallenius',binsize=20, 
                       output.file=tempfile())
                       

## End(Not run)

SCCC-BBC/PathwaySplice documentation built on May 9, 2019, 11:07 a.m.