comparison enrichment_v3.R @ 9:70c0c8757f5f draft

planemo upload commit 9d3e0b226140b566fc529fd0ffe7aa9e8388c6e5-dirty

8:ddaa0c318d65

# OUTPUT :
#	- outputs commanded by the user named respectively result.tsv for the text
#	results file, barplot.png for the barplot image file and dotplot.png for the
#	dotplot image file 


# loading topGO library
library(topGO)

# Read file and return file content as data.frame
readfile = function(filename, header) {
  if (header == "true") {
    # Read only first line of the file as header:

    file <- file[!apply(is.na(file) | file == "", 1, all), , drop=FALSE]
  }
  return(file)
} 

'%!in%' <- function(x,y)!('%in%'(x,y))


# Parse command line arguments

if (length(options.args) != 12) {
    stop("Not enough/Too many arguments", call. = FALSE)
}

typeinput = options.args[1]
listfile = options.args[2]
onto = as.character(options.args[3])
option = as.character(options.args[4])

    sample = readfile(listfile, "true")  
  }else{
    sample = readfile(listfile, "false")
  }
  sample = sample[,column]

}
# Launch enrichment analysis and return result data from the analysis or the null
# object if the enrichment could not be done.
goEnrichment = function(geneuniverse,sample,onto){

	# get all the GO terms of the corresponding ontology (BP/CC/MF) and all their

  geneRatio = data$Significant/data$Annotated
	goTerms = data$Term
	count = data$Significant
  
	labely = paste("GO terms",onto,sep=" ")
	png(filename="dotplot.png",res=300, width = 3200, height = 3200, units = "px")
	sp1 = ggplot(data,aes(x=geneRatio,y=goTerms, color=values,size=count)) +geom_point() + scale_colour_gradientn(colours=c("red","violet","blue")) + xlab("Gene Ratio") + ylab(labely) + labs(color="p-values\n") 

	plot(sp1)
	dev.off()
}

createBarPlot = function(data, onto){

  

  values = roundValues(values)

  values = as.numeric(values)
  goTerms = data$Term
	count = data$Significant
	png(filename="barplot.png",res=300, width = 3200, height = 3200, units = "px")
	
	labely = paste("GO terms",onto,sep=" ")
  p<-ggplot(data, aes(x=goTerms, y=count,fill=values)) + ylab("Gene count") + xlab(labely) +geom_bar(stat="identity") + scale_fill_gradientn(colours=c("red","violet","blue")) + coord_flip() + labs(fill="p-values\n") 
	plot(p)
	dev.off()
}


# Produce the different outputs
createOutputs = function(result, cut_result,text, barplot, dotplot, onto){

	
	if (dotplot=="TRUE"){
	
		createDotPlot(cut_result, onto)
	}
  return(TRUE)
}



# Load R library ggplot2 to plot graphs
library(ggplot2)

# Launch enrichment analysis
allresult = goEnrichment(geneuniverse,sample,onto)
result = allresult[1][[1]]
myGOdata = allresult[2][[1]]
if (!is.null(result)){

	# Adjust the result with a multiple testing correction or not and with the user

9:70c0c8757f5f

# OUTPUT :
#	- outputs commanded by the user named respectively result.tsv for the text
#	results file, barplot.png for the barplot image file and dotplot.png for the
#	dotplot image file 

options(warn=-1)  #TURN OFF WARNINGS !!!!!!

# loading topGO library
suppressMessages(library(topGO))

# Read file and return file content as data.frame
readfile = function(filename, header) {
  if (header == "true") {
    # Read only first line of the file as header:

    file <- file[!apply(is.na(file) | file == "", 1, all), , drop=FALSE]
  }
  return(file)
} 

check_ens_ids <- function(vector) {
  ens_pattern = "^(ENS[A-Z]+[0-9]{11}|[A-Z]{3}[0-9]{3}[A-Za-z](-[A-Za-z])?|CG[0-9]+|[A-Z0-9]+\\.[0-9]+|YM[A-Z][0-9]{3}[a-z][0-9])$"
  return(grepl(ens_pattern,vector))
}

'%!in%' <- function(x,y)!('%in%'(x,y))


# Parse command line arguments

if (length(options.args) != 12) {
    stop("Not enough/Too many arguments", call. = FALSE)
}

#save(options.args,file="/home/dchristiany/proteore_project/ProteoRE/tools/topGO/args.Rda")
#load("/home/dchristiany/proteore_project/ProteoRE/tools/topGO/args.Rda")


typeinput = options.args[1]
listfile = options.args[2]
onto = as.character(options.args[3])
option = as.character(options.args[4])

    sample = readfile(listfile, "true")  
  }else{
    sample = readfile(listfile, "false")
  }
  sample = sample[,column]
}

#check of ENS ids
if (! any(check_ens_ids(sample))){
  print("no ensembl gene ids found in your ids list, please check your IDs in input or the selected column of your input file")
  stop()
}

# Launch enrichment analysis and return result data from the analysis or the null
# object if the enrichment could not be done.
goEnrichment = function(geneuniverse,sample,onto){

	# get all the GO terms of the corresponding ontology (BP/CC/MF) and all their

  geneRatio = data$Significant/data$Annotated
	goTerms = data$Term
	count = data$Significant
  
	labely = paste("GO terms",onto,sep=" ")
	ggplot(data,aes(x=geneRatio,y=goTerms, color=values,size=count)) +geom_point( ) + scale_colour_gradientn(colours=c("red","violet","blue")) + xlab("Gene Ratio") + ylab(labely) + labs(color="p-values\n" ) 
  ggsave("dotplot.png", device = "png", dpi = 320, limitsize = TRUE, width = 15, height = 15, units="cm")
}

createBarPlot = function(data, onto){

  

  values = roundValues(values)

  values = as.numeric(values)
  goTerms = data$Term
	count = data$Significant
	
	labely = paste("GO terms",onto,sep=" ")
  ggplot(data, aes(x=goTerms, y=count,fill=values,scale(scale = 0.5))) + ylab("Gene count") + xlab(labely) +geom_bar(stat="identity") + scale_fill_gradientn(colours=c("red","violet","blue")) + coord_flip() + labs(fill="p-values\n") 
  ggsave("barplot.png", device = "png", dpi = 320, limitsize = TRUE, width = 15, height = 15, units="cm")
}


# Produce the different outputs
createOutputs = function(result, cut_result,text, barplot, dotplot, onto){

	
	if (dotplot=="TRUE"){
	
		createDotPlot(cut_result, onto)
	}
}



# Load R library ggplot2 to plot graphs
suppressMessages(library(ggplot2))

# Launch enrichment analysis
allresult = suppressMessages(goEnrichment(geneuniverse,sample,onto))
result = allresult[1][[1]]
myGOdata = allresult[2][[1]]
if (!is.null(result)){

	# Adjust the result with a multiple testing correction or not and with the user