From b52170d6643c857009f9d66b78e20c4832f4b457 Mon Sep 17 00:00:00 2001
From: DavyCats <davycats.dc@gmail.com>
Date: Thu, 24 May 2018 13:50:58 +0200
Subject: [PATCH] change R script
---
mergecounts.wdl | 24 ++++++++++++++++--------
1 file changed, 16 insertions(+), 8 deletions(-)
diff --git a/mergecounts.wdl b/mergecounts.wdl
index e93068e..8aa2a92 100644
--- a/mergecounts.wdl
+++ b/mergecounts.wdl
@@ -3,8 +3,9 @@ task MergeCounts {
Array[File] inputFiles
String outputFile
- String idVar
- String measurementVar
+ Int featureColumn
+ Int valueColumn
+ Boolean inputHasHeader
# Based on a script by Szymon Kielbasa/Ioannis Moustakas
command <<<
@@ -17,15 +18,22 @@ task MergeCounts {
listOfFiles <- c("${sep='", "' inputFiles}")
+ valueI <- ${valueColumn}
+ featureI <- ${featureColumn}
+ header <- ${true="TRUE" false="FALSE" inputHasHeader}
+
d <- do.call(rbind, lapply(listOfFiles, function(file){
- d <- read.table(file, header=TRUE, comment.char="#")
- colI <- grep(${measurementVar}, colnames(d))
- colnames(d)[colI] <- strsplit(file, "/")[[1]][3]
- d <- d %>% melt(id.vars=${idVar}, measure.vars=colI,
- variable.name="sample", value.name="count")
+ d <- read.table(file, header=header, comment.char="#")
+
+ splitPath <- strsplit(file, "/")[[1]]
+ colnames(d)[valueI] <- sub("\\\.[^\\\.]*$", "",
+ splitPath[length(splitPath)])
+ colnames(d)[featureI] <- "feature"
+
+ d <- d %>% melt(id.vars=featureI, variable.name="sample", value.name="count")
}))
- d <- d %>% dcast(paste0(${idVar}, " ~ sample"), value.var="count")
+ d <- d %>% dcast(feature ~ sample, value.var="count")
write.table(d, sep="\t", quote=FALSE, row.names=FALSE)
CODE
>>>
--
GitLab