Get an input file for color-KEGG-pathways

SPIA_KEGG.color.input.R

##This script get an input file for color-KEGG-pathways
##https://github.com/ajmazurie/color-KEGG-pathways/
##Based from SPIA results at these gists 
##https://gist.github.com/1945349
##https://gist.github.com/1950232

#Load a library needed for text manipulation
#for install run 
#install.packages("stringr")
library("stringr")

#get a link for the pathway you need, in this case the 1st, form SPIA results .
link <- spia_result$KEGGLINK[1]

#Manipulate text to get all info the needed (KEGG_Pathway_ID,KEGG_obj_ID,value)
remove.link <- str_replace(link, "http://www.genome.jp/dbget-bin/show_pathway[?]", "")
texto <- gsub('[+]',',',remove.link)
texto.t <- read.table(textConnection(texto), sep = ",")

#Get all the ENTREZ gene IDs from the SPIA KEGG link
genes.via <- as.numeric(texto.t)
#The last step assign 1 for the pathway name, so we can get rid for this with
genes.via.sel <- c(genes.via[genes.via>1])

#Get the name of the pathway
via <- texto.t[,1]

#Get the name of the organism or KEGG object
org <- str_extract(via, "[a-z]+")

#Get a subset of logFC values with pathway linked genes
genes.both <- sig_genes[which(names(sig_genes) %in% genes.via.sel)]
#Load the results as data frame
genes.both.df <- as.data.frame(genes.both)

#KEGG_obj_ID
via.out <- c(paste(org ,names(genes.both), sep=":"))

#KEGG_Pathway_ID
ruta <- rep(via, length(via.out)) 

#Collect all the info needed for the input file as data.frame
color.input <- data.frame(KEGG_Pathway_ID=ruta, KEGG_obj_ID=via.out, value=genes.both.df[[1]])

#Write the input as csv file.
write.table(color.input, paste(via, "color.input.csv", sep="."), sep=",", quote=FALSE, row.names = FALSE,
            col.names = FALSE)