more script consolidations

1 files changed, 28 insertions, 6 deletions

diff --git a/scripts/pa-groups.R b/scripts/pa-groups.R
index ae4a3c3..3c6ce2c 100755
--- a/scripts/pa-groups.R
+++ b/scripts/pa-groups.R
@@ -1,11 +1,33 @@
 #!/usr/bin/env Rscript
 library(ggplot2)
-data <- read.csv("pyrrolizidine-alkaloids/pa-predictions.csv")
+args = commandArgs(trailingOnly=TRUE)
+groups = read.csv(args[1],header=T)
+data = read.csv(args[2])
 for (i in c(2:10)) {
-  name <- names(data)[i]
-  cols <- append(c(15,17),c(19:28))
-  group <- data[data[i] == 1,cols]
-  freq <- 100*colSums(group,na.rm=TRUE)/colSums(!is.na(group))
-  plot <- ggplot(data.frame(freq),aes(x=names(data)[cols],y=freq)) + geom_bar(stat="identity") + ylab("% mutagenic") + xlab(element_blank()) + ylim(c(0,100)) + theme(axis.text.x = element_text(angle=90))
+  name = names(groups)[i]
+  cols = c(2:15)
+  group = data[groups[i] == 1,cols]
+  freq = 100*colSums(group,na.rm=TRUE)/colSums(!is.na(group))
+  algos = toupper(names(data)[cols])
+  algos = gsub("HIGH",'HC',algos)
+  algos = gsub(".CONFIDENCE",'',algos)
+  algos = gsub("\\.",'-',algos)
+  algos <- factor(algos,levels=rev(c(
+  "MP2D-LAZAR-ALL",
+  "MP2D-LAZAR-HC",
+  "MP2D-RF",
+  "MP2D-LR",
+  "MP2D-LR2",
+  "MP2D-NN",
+  "MP2D-SVM",
+  "CDK-LAZAR-ALL",
+  "CDK-LAZAR-HC",
+  "CDK-RF",
+  "CDK-LR",
+  "CDK-LR2",
+  "CDK-NN",
+  "CDK-SVM"
+  )))
+  plot = ggplot(data.frame(freq),aes(x=freq,y=algos)) + geom_bar(stat="identity") + xlab("% mutagenic") + ylab(element_blank()) + xlim(c(0,100))# + theme(axis.text.x = element_text(angle=90))
   ggsave(paste("figures/",name,".png",sep=""))
 }