vignettes/iedge-tcga-example.Rmd
iedge-tcga-example.Rmd
In this tutorial, we run iEDGE on the TCGA PAAD dataset. To achieve reasonable run time, we restrict the run to the first 3 copy number alterations in the dataset.
For this tutorial, we use a processed iEDGE object. See iEDGE analysis of a simulated dataset for a description of the object’s components. Also, see scripts_TCGA_pancancer/1.preprocess_all_TCGA.R in the vignette folder for an illustration of how to datasets were preprocessed for the TCGA pancancer analysis.
f_in <- "./tcga_sample_data/TCGA_PAAD.RDS"
header <- paste(gsub(".RDS", "", basename(f_in)), sep = "")
cat(paste("header: ", header, "\n", sep = ""))
dat <- readRDS(f_in)
# subset dat to first 3 copy number alterations
names(dat)
cn <- dat$cn[1:3,]
gep <- dat$gep
cisgenes <- dat$cisgenes[1:3]
dat <- list(cn = cn, gep = gep, cisgenes = cisgenes)
# read in geneset annotation files for pathway enrichment analysis
gs.dir <- "./tcga_sample_data"
gs.names <- c("h.all.v5.0.symbols.gmt","c2.cp.v5.0.symbols.gmt", "c3.tft.v5.0.symbols.gmt")
gs.names <- paste(gs.dir, gs.names, sep = "/")
# directory for output reports
f_out <- "."
res <- run_iEDGE(dat, header, f_out,
gs.file = gs.names,
gepid = "gene_symbol",
cnid = "Unique.Name",
cndir = "alteration_direction",
fdr.cis.cutoff = 0.25,
fdr.trans.cutoff = 0.01,
fc.cis = 1.2,
fc.trans = 1.5,
min.drawsize = 3,
onesided.cis = TRUE,
onesided.trans = FALSE,
uptest = "Amplification",
downtest = "Deletion",
min.group = 2,
prune.col = "fdr",
prune.thres = 0.05,
hyperthres = 0.25,
cis.boxplot = TRUE,
trans.boxplot = FALSE,
bipartite = TRUE,
enrichment = TRUE,
html = TRUE)
The HTML report of this run is contained here