You are currently viewing the SEQanswers forums as a guest, which limits your access. Click here to register now, and join the discussion
). However, on the advice of one of the cummeRbund authors, Loyal, I was able to make log fold-change heat maps by extracting raw FPKM values from fpkmMatrix(), do the log2FC transformation in R, then use heatmap.2() from gplots to create something like the attached file. ). However, on the advice of one of the cummeRbund authors, Loyal, I was able to make log fold-change heat maps by extracting raw FPKM values from fpkmMatrix(), do the log2FC transformation in R, then use heatmap.2() from gplots to create something like the attached file.
library(cummeRbund); library(gplots)
cuff <- readCufflinks("cuffdiff_out",rebuild=T)
db <- fpkmMatrix(genes(cuff))
sigGenes <- getSig(cuff,'BY4741','ino80',alpha=0.05,level='genes')
db <- db[sigGenes,]
WT <- "BY4741" # Set the column name of the WT sample
mut <- c("ino80","arp5","ies6") # Set column names of test conditions you want in the analysis
logFC <- function(db,mutants,WT,logBase=2,pseudo=1) {
if (length(WT) !=1 ) {
stop('WT must refer to a single gene/column')
}
if (is.numeric(logBase)==FALSE) {
stop('logBase must be a numeric value')
}
db <- (db[,mutants]+pseudo)/(db[,WT]+pseudo)
db <- log(db,logBase)
}
db <- logFC(db,mutants=mut,WT=WT,logBase=2,pseudo=1) # This does the log transformation
db <- as.matrix(db)
heatmap.2(
db,
Rowv=TRUE,
Colv=FALSE,
dendrogram="row",
trace="none",
labRow="",
density.info=c("none"),
main="Log(Fold-Change) \nExpression Profiles"
)
). However, on the advice of one of the cummeRbund authors, Loyal, I was able to make log fold-change heat maps by extracting raw FPKM values from fpkmMatrix(), do the log2FC transformation in R, then use heatmap.2() from gplots to create something like the attached file. ). However, on the advice of one of the cummeRbund authors, Loyal, I was able to make log fold-change heat maps by extracting raw FPKM values from fpkmMatrix(), do the log2FC transformation in R, then use heatmap.2() from gplots to create something like the attached file.
Leave a comment: