add replicas to plot

benchmark/plots.r

@@ -6,6 +6,11 @@ library(dplyr)
 library(forcats)
 
 df <- read.csv("/home/mcroiset/HiC/benchmark/recap.txt", sep = "\t", header = TRUE)
+rep2 <- read.csv("/home/mcroiset/HiC/benchmark/recap_rep2.txt", sep = "\t", header = TRUE)
+rep3 <- read.csv("/home/mcroiset/HiC/benchmark/recap_rep3.txt", sep = "\t", header = TRUE)
+
+merge_df <- bind_rows(df, rep2, rep3)
+
 
 convert_time <- function(x){
   parts <- strsplit(x, " ")
@@ -45,44 +50,44 @@ convert_time <- function(x){
   return(list_durations)
 }
 
-list_dur <- convert_time(df$duration)
-list_realtime <- convert_time(df$realtime)
+list_dur <- convert_time(merge_df$duration)
+list_realtime <- convert_time(merge_df$realtime)
 
-df$duration_minutes <- list_dur
-df$duration_minutes <- as.numeric(df$duration_minutes)
+merge_df$duration_minutes <- list_dur
+merge_df$duration_minutes <- as.numeric(merge_df$duration_minutes)
 
-df$realtime_minutes <- list_realtime
-df$realtime_minutes <- as.numeric(df$realtime_minutes)
+merge_df$realtime_minutes <- list_realtime
+merge_df$realtime_minutes <- as.numeric(merge_df$realtime_minutes)
 
-val <- df %>% group_by(name) %>% summarise(moy = mean(duration_minutes), ect = sd(duration_minutes))
+val <- merge_df %>% group_by(name) %>% summarise(moy = mean(duration_minutes), ect = sd(duration_minutes))
 
 workflow <- character(0)
-for (x in df$file) {
+for (x in merge_df$file) {
   full <- sapply(strsplit(x, split = "/"), "[", 2)
   w <- sapply(strsplit(full, split = "_"), "[", 1)
   workflow <- append(workflow,w)
 }
-df$workflow <- workflow
+merge_df$workflow <- workflow
 
 processes <- character(0)
-for (x in df$name) {
+for (x in merge_df$name) {
   process <- lapply(strsplit(x, split = ":"), tail, n = 1L)  
   processes <- append(processes, as.character(process))
 }
-df$name <- processes
+merge_df$name <- processes
 
 filenames <- character(0)
-for (x in df$file) {
+for (x in merge_df$file) {
   file_short_name <- sapply(strsplit(x, split = "/"), "[", 2)
   filenames <- append(filenames,file_short_name)
 }
-df$file <- filenames
+merge_df$file <- filenames
 
 important_processes <- c(15, 16, 17, 18, 7, 1, 1, 1, 1, 1, 1, 1, 1,1,1,1,1,9,1,2,5,6,1,1,1,1,1,15,11,1,1,1,1,12,1,1,1,1,1)
 
 categories <- character(0)
-for (x in df$name) {
-  if (grepl('\\w*_ALIGN\\w*', x) || grepl('\\w*TRIM\\w*', x) || grepl('MERGE_BOWTIE2', x)) {
+for (x in merge_df$name) {
+  if (grepl('\\w*ALIGN\\w*', x) || grepl('\\w*TRIM\\w*', x) || grepl('MERGE_BOWTIE2', x)) {
     categories <- append(categories, "align")
   }
   else if (grepl('\\w*COOLER\\w*', x)) {
@@ -91,7 +96,7 @@ for (x in df$name) {
   else if (grepl('\\w*PAIRS\\w*', x)) {
     categories<- append(categories, "pairs")
   }
-  else if (grepl('\\w*MATRIX\\w*', x)) {
+  else if (grepl('\\w*MATRIX\\w*', x) || grepl('\\w*BUILD_CONTACT\\w*', x) || grepl('\\w*ICE_NORM\\w*', x)) {
     categories<- append(categories, "matrix")
   }
   else if (grepl('\\w*FILTER\\w*', x) || grepl('\\w*PICARD\\w*', x) || grepl('\\w*SAMTOOLS\\w*', x)) {
@@ -100,9 +105,6 @@ for (x in df$name) {
   else if (grepl('\\w*CUTSITE\\w*', x)) {
     categories<- append(categories, "cutsite")
   }
-  else if (grepl('\\w*ITERALIGN\\w*', x)) {
-    categories<- append(categories, "iteralign")
-  }
   else if (grepl('\\w*SAMPLESHEET\\w*', x) || (grepl('BOWTIE2_BUILD', x) || grepl('\\w*GETCHROM\\w*', x) || grepl('\\w*GET_RESTRIC\\w*', x))) {
     categories<- append(categories, "data_prep")
   }
@@ -111,21 +113,21 @@ for (x in df$name) {
   }
 }
 # categories
-df$categorie <- categories
+merge_df$categorie <- categories
 
-print(levels(as.factor(df$name)))
-ordered_processes <- tibble(levels(as.factor(df$name)))
-order <- c(18, 7, 8, 7, 2, 20, 21, 14, 24, 23, 27, 22, 25, 30, 31, 32, 3, 6, 5, 19, 13, 19, 9, 4, 15, 28, 29, 18, 7, 8, 17, 16, 33, 11, 1, 10, 12, 27, 8)
+#print(levels(as.factor(merge_df$name)))
+ordered_processes <- tibble(levels(as.factor(merge_df$name)))
+order <- c(18, 7, 8, 7, 2, 20, 20, 21, 14, 24, 23, 27, 22, 25, 30, 31, 32, 3, 6, 5, 19, 13, 19, 9, 4, 15, 28, 29, 18, 21, 7, 8, 17, 16, 33, 11, 1, 10, 12, 27, 8)
 ordered_processes <- ordered_processes %>% add_column(order = order)
-ordered_processes <- rename(ordered_processes, name = `levels(as.factor(df$name))`)
+ordered_processes <- rename(ordered_processes, name = `levels(as.factor(merge_df$name))`)
 
-df <- left_join(df, ordered_processes, by= c("name" = "name"))
+merge_df <- left_join(merge_df, ordered_processes, by= c("name" = "name"))
 
 
 
-# pdf("process_time.pdf")
+h# pdf("process_time.pdf")
 
-ggplot(df, aes(x = name, y = duration_minutes)) +
+ggplot(merge_df, aes(x = name, y = duration_minutes)) +
   geom_point(aes(color = name, shape = name)) +
   scale_shape_manual(values = important_processes) +
   facet_wrap(~ file, ncol = 8) +
@@ -135,15 +137,18 @@ ggplot(df, aes(x = name, y = duration_minutes)) +
   theme(axis.text.x=element_blank(), axis.ticks.x=element_blank()) + 
   ggtitle("Duration time of each process")
 
-ggplot(df, aes(x = fct_reorder(name, order), y = realtime_minutes)) +
-  geom_point(aes(color = categorie)) +
+ggplot(merge_df, aes(x = fct_reorder(name, order), y = realtime_minutes)) +
+  geom_boxplot(aes(color = categorie)) +
   scale_shape_manual(values = important_processes) +
   facet_wrap(~ file, ncol = 8) +
   xlab("Processes") +
   ylab("Duration in minutes") +
   theme(axis.text.x=element_blank(), axis.ticks.x=element_blank())+ 
   ggtitle("Real execution time of each process") +
-  scale_y_log10()
+  scale_y_log10() +
+  theme_bw()
 
 
 # dev.off()
+
+merge_df