compareResults: compareResults

Description Usage Arguments Value Examples

Description

This function helps with visualizing the effects of bias adjustment in pathway analysis, by comparing the distributions of bias factors (e.g. number of exon bins) in genes associated with the most significant gene sets, before and after adjusting for bias factors in splicing pathway analysis.

Usage

1
2
compareResults(n.go, adjusted, unadjusted, gene.based.table,
  output.dir = tempdir(), type.boxplot = c("All", "Only3"))

Arguments

n.go

Distributions of bias factor in genes associated with the most significant n.go gene sets will be compared

adjusted

An object returned by runPathwaySplice, should correspond to gene set anlaysis results adjusting for biases in splicing analysis

unadjusted

An object returned by runPathwaySplice, should correspond to gene set analysis results NOT adjusting for biases

gene.based.table

An object returned by makeGeneTable, should correspond to a table with one p-value for each gene

output.dir

Directory for output files

type.boxplot

Options are 'All' and 'Only3', corresponding to drawing 5 boxplots or 3 boxplots.

5 boxplots: all genesets, sig.adjusted (sig gene sets in adjusted analysis), sig.unadjusted (sig gene sets in unadjusted analysis), sig.adjusted.only (sig gene sets in adjusted analysis only), sig.unadjusted.only (sig gene sets in unadjusted analysis only)

3 boxplots: all genesets, adjusted.sig, unadjusted.sig

Value

The output include 3 files in output.dir: (1) a venn diagram comparing significant gene sets before and after adjusting for bias factors (2) a .csv file with gene set names belonging to different sections of the venn diagram (3) a box plot showing the distributions of number of features within all genes in significant gene sets, with or without adjusting for bias factors

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
dir.name <- system.file('extdata', package='PathwaySplice')
hallmark.pathway.file <- file.path(dir.name,'h.all.v6.0.symbols.gmt.txt')

hallmark <- gmtGene2Cat(hallmark.pathway.file,genomeID='hg19')
                   
gene.based.table <- makeGeneTable(featureBasedData)

res.adj <- runPathwaySplice(gene.based.table,genome='hg19',
                         id='ensGene',gene2cat=hallmark,  
                         go.size.limit = c(5, 200),
                         method='Wallenius')

res.unadj <- runPathwaySplice(gene.based.table,genome='hg19',
                         id='ensGene',gene2cat=hallmark,go.size.limit = c(5, 200),
                         method='Hypergeometric')

compareResults(20, res.adj, res.unadj, gene.based.table, type.boxplot='Only3')

## Not run: 
# illustrate specification of output directory on windows systems
compareResults(20, res.adj, res.unadj, gene.based.table, type.boxplot='Only3',output.dir=tempdir())

output.dir <- '~/OutputTestPathwaySplice' #linux system
compareResults(20,res.adj, res.unadj,gene.based.table, output.dir, type.boxplot='Only3')

## End(Not run)

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