STAR_WASP_Bechmarking_Analysis

2.1 Loading and preprocessing benchmark data from all runs - data extracts from the system-time command

#run_results <- read.table(pipe ('ssh atlas "cat /home/asiimwe/projects/run_env/alpha_star_wasp_comparison/benchmark_results_all_runs.txt"'), sep="\t", header = FALSE, fill=TRUE, quote = "") #remote call
run_results <- read.table("benchmark_results_all_runs.txt", sep="\t", header = FALSE, fill=TRUE, quote = "")

run_results <- as.data.frame(run_results[-c(1),])
run_results$V4 <- NULL
colnames(run_results) <- c("Sample", "Run", "Thread", "Param","Value")

run_results$Param <- as.character(run_results$Param)
run_results$Param[run_results$Param == "Elapsed (wall clock) time (h"] <- "Wall_Clock"
run_results$Param[run_results$Param == "Maximum resident set size (kbytes)"] <- "Memory" 

run_results$Value <- as.character(run_results$Value)
run_results$Value <- gsub("mm:ss or m:ss): ", "", run_results$Value)
run_results$Value <- gsub(" ", "", run_results$Value)
run_results$Value

unique(run_results$Run)
run_results_STAR <- run_results %>% filter(Run == "STAR_Runs_noVars_noWaspOutput")#STAR base - no variants
run_results_STAR_WASP <- run_results %>% filter(Run == "STAR_Runs") #STAR WASP
run_results_WASP <- run_results %>% filter(Run == "STAR_WASP_Runs") #WASP - change naming globally 
run_results_STAR$Run <- NULL
run_results_STAR_WASP$Run <- NULL
run_results_WASP$Run <- NULL

run_results_STAR <- spread(run_results_STAR, key = Thread, value = Value)
colnames(run_results_STAR) <- c("Sample", "Param", "16 Threads-STAR", "32 Threads-STAR", "8 Threads-STAR")

run_results_STAR_WASP <- spread(run_results_STAR_WASP, key = Thread, value = Value)
colnames(run_results_STAR_WASP) <- c("Sample", "Param", "16 Threads-STAR_WASP", "32 Threads-STAR_WASP", "8 Threads-STAR_WASP")

run_results_WASP <- spread(run_results_WASP, key = Thread, value = Value)
colnames(run_results_WASP) <- c("Sample", "Param", "16 Threads-WASP", "32 Threads-WASP", "8 Threads-WASP")

data <- merge(run_results_STAR, run_results_STAR_WASP, by = c("Sample", "Param"))
data <- merge(data, run_results_WASP, by = c("Sample", "Param"))
#write.csv(data, file = "/home/asiimwe/projects/run_env/alpha_star_wasp_comparison/benchmark_results_all_runs_pivoted.csv")

#Plotting speeds and memory
data_melted <- melt(data, id.vars=c("Sample", "Param"))
data_melted <- cSplit(data_melted,"variable", "-", direction="wide")

colnames(data_melted) <- c("Sample", "Param", "Value", "Threads", "Run")
data_melted$Param <- as.character(data_melted$Param)

#data_melted_subset_clock <- data_melted_subset %>% filter(Param == "Wall_Clock")
#data_melted_subset_clock$Value <- gsub("\\.", ":", data_melted_subset_clock$Value)
#data_melted_subset_clock$Value
#data_melted_subset_clock$Value  <- chron(times. = data_melted_subset_clock$Value)#rejecting format - expecting h:m:s but third aguments is more than 60 for some instances


###########------------------------------------------------------------------------------
# #compressed dataset - removing parameters that are not needed
# #parameters to remove:
# param_remove <- c(
#   "Average resident set size (kbytes)",
#   "Average shared text size (kbytes)",
#   "Average stack size (kbytes)",
#   "Average total size (kbytes)",
#   "Average unshared data size (kbytes)",
#   "Exit status",
#   "Major (requiring I/O) page faults",
#   "Signals delivered",
#   "Socket messages received",
#   "Socket messages sent",
#   "Swaps"
# )
# 
# dataset_subset <- data %>% filter(Param %nin% param_remove)
# write.csv(dataset_subset, file = "/home/asiimwe/projects/run_env/alpha_star_wasp_comparison/run_results_subset_pivoted.csv")
###########------------------------------------------------------------------------------

#Benchmark plots:
#Elapsed (wall clock) time (h    mm:ss or m:ss):       Average resident set size (kbytes)  

2.2 Loading Log.final.out results for all runs

# Loading Log.final.out results to extract required parameters
final.log.results <- read.table("final_log_results_all_runs.txt", sep="|", header = FALSE, fill=TRUE, quote = "")
#final.log.results <- read.table(pipe('ssh atlas "cat /home/asiimwe/projects/run_env/alpha_star_wasp_comparison/final_log_results_all_runs.txt"'), sep="|", header = FALSE, fill=TRUE, quote = "")

colnames(final.log.results) <- unlist(final.log.results[c(1),])
final.log.results <- as.data.frame(final.log.results[-c(1),])
final.log.results <- as.data.frame(final.log.results)

final.log.results$Sample <- as.character(final.log.results$Sample)
final.log.results$Run <- as.character(final.log.results$Run)
final.log.results$Thread <- as.character(final.log.results$Thread)
final.log.results$Param <- as.character(final.log.results$Param)

colnames(final.log.results)
final.log.results$Param <- str_trim(final.log.results$Param)

final.log.results_filt <- final.log.results %>% filter(Value != "")

final.log.results_filt$Run <- as.character(final.log.results_filt$Run)
unique(final.log.results_filt$Run)
final.log.results_filt$Run[final.log.results_filt$Run == "STAR_Runs_noVars_noWaspOutput"] <- "STAR"
final.log.results_filt$Run[final.log.results_filt$Run == "STAR_Runs"] <- "STAR_WASP"
final.log.results_filt$Run[final.log.results_filt$Run == "STAR_WASP_Runs"] <- "WASP"
colnames(final.log.results_filt)[3] <- "Threads"


#Extracting number of reads 
final.log.results_filt_input_reads <- final.log.results_filt %>% filter(Param == "Number of input reads") 
final.log.results_filt_mapping_speed <- final.log.results_filt %>% filter(Param == "Mapping speed, Million of reads per hour") 
final.log.results_filt_average_input_read_length <- final.log.results_filt %>% filter(Param == "Average input read length")


data_melted_join_pass1 <- inner_join(data_melted, final.log.results_filt_input_reads, by = c("Sample", "Run", "Threads"))
data_melted_join_pass2 <- inner_join(data_melted_join_pass1, final.log.results_filt_mapping_speed, by = c("Sample", "Run", "Threads"))
data_melted_beta <- inner_join(data_melted_join_pass2, final.log.results_filt_average_input_read_length, by = c("Sample", "Run", "Threads"))

colnames(data_melted_beta)
colnames(data_melted_beta) <- c( "Sample", "Param", "Value", "Threads", "Run", "Param.y", "Number_of_input_reads", "Param.x.x", "Mapping_speed_Million_of_reads_per_hour", "Param.y.y", "Average_input_read_length")
data_melted_beta$Param.y <- NULL
data_melted_beta$Param.x.x <- NULL
data_melted_beta$Param.y.y <- NULL 


nrow(data_melted)
nrow(final.log.results_filt_input_reads)
nrow(final.log.results_filt_mapping_speed)
nrow(final.log.results_filt_average_input_read_length)
nrow(data_melted_beta)
#write.csv(data_melted_beta, file = "/home/asiimwe/projects/run_env/alpha_star_wasp_comparison/data_melted_beta.csv")

## Preparing data for Latex tables

# Installing required packages
# tinytex::parse_install(
#   text = "! LaTeX Error: File `ae.sty' not found."
# )
# 
# 
# tinytex::parse_install(
#   text = "! LaTeX Error: File `fullpage.sty' not found."
# )
# tinytex::parse_install(
#   text = "! LaTeX Error: File `moresize.sty' not found."
# )
# 
# 
# tinytex::parse_install(
#   text = "! LaTeX Error: File `colortbl.sty' not found."
# )
# 
# 
# tinytex::parse_install(
#   text = "! LaTeX Error: File `soul.sty' not found."
# )
# 
# tinytex::parse_install(
#   text = "! LaTeX Error: File `adjustbox.sty' not found."
# )
# 
# 
# tinytex::parse_install(
#   text = "! LaTeX Error: File `collectbox.sty' not found."
# )
# tinytex::parse_install(
#  text = "! LaTeX Error: File `pdflscape.sty' not found."
#  )
#tinytex::parse_install(
#   text = "! LaTeX Error: File `grfext.sty' not found."
# )
# tinytex::parse_install(
#    text = "! LaTeX Error: File `tabu.sty' not found."
#  )
# 
# tinytex::parse_install(
#    text = "! LaTeX Error: File `threeparttable.sty' not found."
#  )
#data_melted_beta <- read.csv("/home/asiimwe/projects/run_env/alpha_star_wasp_comparison/data_melted_beta.csv")

data_melted_beta
data_melted_beta$Threads <- ordered(data_melted_beta$Threads , levels = c("8 Threads", "16 Threads", "32 Threads"))
unique(data_melted_beta$Param)
data_melted_beta_subset <- data_melted_beta %>% filter(Param == "Wall_Clock" | Param == "Memory")
data_melted_beta_subset$Average_input_read_length <- NULL
#data_melted_beta_subset$threads_run <- paste0(data_melted_beta_subset$Threads, "_", data_melted_beta_subset$Run)
#data_melted_beta_subset$Param <- NULL
#data_melted_beta_subset$Threads <- NULL
#data_melted_beta_subset$Run <- NULL

#-----------------------STAR-----------------------
data_melted_beta_subset_STAR <- data_melted_beta_subset %>% filter(Run == "STAR")
data_melted_beta_subset_STAR_Param <- data_melted_beta_subset_STAR[,c(1:4)] #check environment differences
data_melted_beta_subset_STAR_reads <- unique(data_melted_beta_subset_STAR[,c(1,6)])
data_melted_beta_subset_STAR_speed <- unique(data_melted_beta_subset_STAR[,c(1,4, 7)])
data_melted_beta_subset_STAR_Param$comb <- paste0(data_melted_beta_subset_STAR_Param$Param, "_", data_melted_beta_subset_STAR_Param$Threads, "_", "STAR")
data_melted_beta_subset_STAR_Param$Param <- NULL
data_melted_beta_subset_STAR_Param$Threads <- NULL
data_melted_beta_subset_STAR_Param <- as.data.frame(data_melted_beta_subset_STAR_Param)
data_melted_beta_subset_STAR_Param_spread <- spread(data_melted_beta_subset_STAR_Param, key = comb, value = Value)

data_melted_beta_subset_STAR_Param_spread <- data_melted_beta_subset_STAR_Param_spread %>% dplyr::group_by(Sample) %>% 
  dplyr::summarise_all(purrr::discard, is.na)

data_melted_beta_subset_STAR_Param_spread <- as.data.frame(data_melted_beta_subset_STAR_Param_spread)

data_melted_beta_subset_STAR_speed_spread <- spread(data_melted_beta_subset_STAR_speed, key = Threads, value = Mapping_speed_Million_of_reads_per_hour)
colnames(data_melted_beta_subset_STAR_speed_spread)
colnames(data_melted_beta_subset_STAR_speed_spread) <- c("Sample", "Mapping_Speed_8 Threads_STAR",  "Mapping_Speed_16 Threads_STAR", "Mapping_Speed_32 Threads_STAR")
STAR_tab_subset <- inner_join(data_melted_beta_subset_STAR_reads, data_melted_beta_subset_STAR_Param_spread, by = "Sample")
STAR_tab_subset <- inner_join(STAR_tab_subset, data_melted_beta_subset_STAR_speed_spread, by = "Sample")
STAR_tab_subset <- STAR_tab_subset[,c(1,2,3,4,5,9,10,11, 6,7,8)]

#----------------------STAR_WASP-----------------------
data_melted_beta_subset_STAR_WASP <- data_melted_beta_subset %>% filter(Run == "STAR_WASP")
data_melted_beta_subset_STAR_WASP_Param <- data_melted_beta_subset_STAR_WASP[,c(1:4)]
data_melted_beta_subset_STAR_WASP_reads <- unique(data_melted_beta_subset_STAR_WASP[,c(1,6)])
data_melted_beta_subset_STAR_WASP_speed <- unique(data_melted_beta_subset_STAR_WASP[,c(1,4, 7)])
data_melted_beta_subset_STAR_WASP_Param$comb <- paste0(data_melted_beta_subset_STAR_WASP_Param$Param, "_", data_melted_beta_subset_STAR_WASP_Param$Threads, "_", "STAR_WASP")
data_melted_beta_subset_STAR_WASP_Param$Param <- NULL
data_melted_beta_subset_STAR_WASP_Param$Threads <- NULL
data_melted_beta_subset_STAR_WASP_Param <- as.data.frame(data_melted_beta_subset_STAR_WASP_Param)
#rownames(data_melted_beta_subset_STAR_WASP_Param) <- 1:nrow(data_melted_beta_subset_STAR_WASP_Param)
#data_melted_beta_subset_STAR_WASP_Param$id <- 1:nrow(data_melted_beta_subset_STAR_WASP_Param)
data_melted_beta_subset_STAR_WASP_Param_spread <- spread(data_melted_beta_subset_STAR_WASP_Param, key = comb, value = Value)
data_melted_beta_subset_STAR_WASP_Param_spread$id <- NULL

data_melted_beta_subset_STAR_WASP_Param_spread <- data_melted_beta_subset_STAR_WASP_Param_spread %>% dplyr::group_by(Sample) %>% 
  dplyr::summarise_all(purrr::discard, is.na)

data_melted_beta_subset_STAR_WASP_Param_spread <- as.data.frame(data_melted_beta_subset_STAR_WASP_Param_spread)

data_melted_beta_subset_STAR_WASP_speed_spread <- spread(data_melted_beta_subset_STAR_WASP_speed, key = Threads, value = Mapping_speed_Million_of_reads_per_hour)
colnames(data_melted_beta_subset_STAR_WASP_speed_spread)
colnames(data_melted_beta_subset_STAR_WASP_speed_spread) <- c("Sample", "Mapping_Speed_8 Threads_STAR_WASP",  "Mapping_Speed_16 Threads_STAR_WASP", "Mapping_Speed_32 Threads_STAR_WASP")
STAR_WASP_tab_subset <- inner_join(data_melted_beta_subset_STAR_WASP_Param_spread, data_melted_beta_subset_STAR_WASP_speed_spread, by = "Sample")
STAR_WASP_tab_subset <- STAR_WASP_tab_subset[,c(1,2,3,4,8,9,10, 5,6,7)]


#-------------------WASP-------------------------------
data_melted_beta_subset_WASP <- data_melted_beta_subset %>% filter(Run == "WASP")
data_melted_beta_subset_WASP_Param <- data_melted_beta_subset_WASP[,c(1:4)]
data_melted_beta_subset_WASP_reads <- unique(data_melted_beta_subset_WASP[,c(1,6)])
data_melted_beta_subset_WASP_speed <- unique(data_melted_beta_subset_WASP[,c(1,4, 7)])
data_melted_beta_subset_WASP_Param$comb <- paste0(data_melted_beta_subset_WASP_Param$Param, "_", data_melted_beta_subset_WASP_Param$Threads, "_", "WASP")
data_melted_beta_subset_WASP_Param$Param <- NULL
data_melted_beta_subset_WASP_Param$Threads <- NULL
data_melted_beta_subset_WASP_Param <- as.data.frame(data_melted_beta_subset_WASP_Param)
#rownames(data_melted_beta_subset_WASP_Param) <- 1:nrow(data_melted_beta_subset_WASP_Param)
#data_melted_beta_subset_WASP_Param$id <- 1:nrow(data_melted_beta_subset_WASP_Param)
data_melted_beta_subset_WASP_Param_spread <- spread(data_melted_beta_subset_WASP_Param, key = comb, value = Value)
data_melted_beta_subset_WASP_Param_spread$id <- NULL

data_melted_beta_subset_WASP_Param_spread <- data_melted_beta_subset_WASP_Param_spread %>% dplyr::group_by(Sample) %>% 
  dplyr::summarise_all(purrr::discard, is.na)

data_melted_beta_subset_WASP_Param_spread <- as.data.frame(data_melted_beta_subset_WASP_Param_spread)

data_melted_beta_subset_WASP_speed_spread <- spread(data_melted_beta_subset_WASP_speed, key = Threads, value = Mapping_speed_Million_of_reads_per_hour)
colnames(data_melted_beta_subset_WASP_speed_spread)
colnames(data_melted_beta_subset_WASP_speed_spread) <- c("Sample", "Mapping_Speed_8 Threads_WASP",  "Mapping_Speed_16 Threads_WASP", "Mapping_Speed_32 Threads_WASP")

WASP_tab_subset <- inner_join(data_melted_beta_subset_WASP_Param_spread, data_melted_beta_subset_WASP_speed_spread, by = "Sample")
WASP_tab_subset <- WASP_tab_subset[,c(1,2,3,4,8,9,10, 5,6,7)]

STAR_tab_subset <- as.data.frame(STAR_tab_subset)
STAR_WASP_tab_subset <- as.data.frame(STAR_WASP_tab_subset)
WASP_tab_subset <- as.data.frame(WASP_tab_subset)

n_way_merger0 <- merge(STAR_tab_subset, STAR_WASP_tab_subset, by = "Sample")
n_way_merger <- merge(n_way_merger0, WASP_tab_subset, by = "Sample")

#n_way_merger$Size_holder1 <- ""
#n_way_merger$Size_holder2 <- ""

colnames(n_way_merger)

n_way_merger <- n_way_merger[, c("Sample", "Number_of_input_reads",   #"Sample", "Size_holder1","Size_holder2", "Number_of_input_reads", 
                                 "Memory_8 Threads_STAR","Memory_16 Threads_STAR", "Memory_32 Threads_STAR",
                                 "Mapping_Speed_8 Threads_STAR","Mapping_Speed_16 Threads_STAR", "Mapping_Speed_32 Threads_STAR", 
                                 "Wall_Clock_8 Threads_STAR","Wall_Clock_16 Threads_STAR", "Wall_Clock_32 Threads_STAR",
                                 "Memory_8 Threads_STAR_WASP", "Memory_16 Threads_STAR_WASP", "Memory_32 Threads_STAR_WASP", 
                                 "Mapping_Speed_8 Threads_STAR_WASP", "Mapping_Speed_16 Threads_STAR_WASP", "Mapping_Speed_32 Threads_STAR_WASP",
                                 "Wall_Clock_8 Threads_STAR_WASP", "Wall_Clock_16 Threads_STAR_WASP", "Wall_Clock_32 Threads_STAR_WASP",
                                 "Memory_8 Threads_WASP", "Memory_16 Threads_WASP", "Memory_32 Threads_WASP", 
                                 "Mapping_Speed_8 Threads_WASP", "Mapping_Speed_16 Threads_WASP", "Mapping_Speed_32 Threads_WASP",
                                 "Wall_Clock_8 Threads_WASP", "Wall_Clock_16 Threads_WASP", "Wall_Clock_32 Threads_WASP"
                                 )] 

#knitr::kable(n_way_merger, "latex")

# Smoothing plot function
geom_xspline <- function(mapping = NULL, data = NULL, stat = "xspline",
                      position = "identity", show.legend = NA,
                      inherit.aes = TRUE, na.rm = TRUE,
                      spline_shape=-0.25, open=TRUE, rep_ends=TRUE, ...) {
  layer(
    geom = GeomXspline,
    mapping = mapping,
    data = data,
    stat = stat,
    position = position,
    show.legend = show.legend,
    inherit.aes = inherit.aes,
    params = list(spline_shape=spline_shape,
                  open=open,
                  rep_ends=rep_ends,
                  ...)
  )
}
 
GeomXspline <- ggproto("GeomXspline", GeomLine,
  required_aes = c("x", "y"),
  default_aes = aes(colour = "black", size = 0.5, linetype = 1, alpha = NA)
)
 
stat_xspline <- function(mapping = NULL, data = NULL, geom = "line",
                     position = "identity", show.legend = NA, inherit.aes = TRUE,
                     spline_shape=-0.25, open=TRUE, rep_ends=TRUE, ...) {
  layer(
    stat = StatXspline,
    data = data,
    mapping = mapping,
    geom = geom,
    position = position,
    show.legend = show.legend,
    inherit.aes = inherit.aes,
    params = list(spline_shape=spline_shape,
                  open=open,
                  rep_ends=rep_ends,
                  ...
    )
  )
}
 
StatXspline <- ggproto("StatXspline", Stat,
 
  required_aes = c("x", "y"),
 
  compute_group = function(self, data, scales, params,
                           spline_shape=-0.25, open=TRUE, rep_ends=TRUE) {
    tf <- tempfile(fileext=".png")
    png(tf)
    plot.new()
    tmp <- xspline(data$x, data$y, spline_shape, open, rep_ends, draw=FALSE, NA, NA)
    invisible(dev.off())
    unlink(tf)
 
    data.frame(x=tmp$x, y=tmp$y)
  }
)

3.1 Memory

#Loading pre-processed data from final log results (all runs)
data_melted_beta <- read.csv("data_melted_beta.csv")
#data_melted_beta <- read.csv(pipe('ssh atlas "cat /home/asiimwe/projects/run_env/alpha_star_wasp_comparison/data_melted_beta.csv"'))

data_melted_beta$X <- NULL
colnames(data_melted_beta)

## [1] "Sample"                                 
## [2] "Param"                                  
## [3] "Value"                                  
## [4] "Threads"                                
## [5] "Run"                                    
## [6] "Number_of_input_reads"                  
## [7] "Mapping_speed_Million_of_reads_per_hour"
## [8] "Average_input_read_length"

#"Sample"  "Param"  "Value" "Threads" "Run" "Mapping_speed_Million_of_reads_per_hour" "Average_input_read_length" 

#Filtering samples not needed downstream:
unique(data_melted_beta$Sample)

##  [1] HG00512                      HG00513                     
##  [3] HG00731                      HG00732                     
##  [5] HG00733                      NA12878_Nucleus_nonPolyA    
##  [7] NA12878_Nucleus_nonPolyA_Rep NA12878_Nucleus_PolyA       
##  [9] NA12878_Nucleus_PolyA_Rep    NA12878_PolyA               
## [11] NA12878_PolyA_Rep            NA12878_RAMPAGE             
## [13] NA12878_RAMPAGE_Rep          NA12878_Small               
## [15] NA12878_Total                NA12878_Total_Rep           
## [17] NA19238                      NA19239                     
## [19] NA19240                     
## 19 Levels: HG00512 HG00513 HG00731 HG00732 HG00733 ... NA19240

data_melted_beta <- data_melted_beta %>% filter(Sample != "NA12878_Small" & Sample != "NA12878_RAMPAGE" & Sample != "NA12878_RAMPAGE_Rep") 

data_melted_beta$Threads <- as.character(data_melted_beta$Threads )
data_melted_beta$Threads[data_melted_beta$Threads == "8threads" ] <- "8 Threads"
data_melted_beta$Threads[data_melted_beta$Threads == "16threads" ] <- "16 Threads"
data_melted_beta$Threads[data_melted_beta$Threads == "32threads" ] <- "32 Threads"
unique(data_melted_beta$Threads)

## [1] "16 Threads" "32 Threads" "8 Threads"

#Data cleanup
str(data_melted_beta)

## 'data.frame':    3168 obs. of  8 variables:
##  $ Sample                                 : Factor w/ 19 levels "HG00512","HG00513",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ Param                                  : Factor w/ 22 levels "Average resident set size (kbytes)",..: 1 2 3 4 5 6 7 8 9 10 ...
##  $ Value                                  : Factor w/ 1638 levels "0","1","1:03:19",..: 1 1 1 1 1 1 1443 1450 663 1 ...
##  $ Threads                                : chr  "16 Threads" "16 Threads" "16 Threads" "16 Threads" ...
##  $ Run                                    : Factor w/ 3 levels "STAR","STAR_WASP",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ Number_of_input_reads                  : int  66055710 66055710 66055710 66055710 66055710 66055710 66055710 66055710 66055710 66055710 ...
##  $ Mapping_speed_Million_of_reads_per_hour: num  1219 1219 1219 1219 1219 ...
##  $ Average_input_read_length              : int  102 102 102 102 102 102 102 102 102 102 ...

data_melted_beta$Sample <- as.character(data_melted_beta$Sample)
data_melted_beta$Param <- as.character(data_melted_beta$Param)
data_melted_beta$Value <- gsub("%", "", data_melted_beta$Value)

#Reordering threads to appear in the correct order (8, 16, 32)
data_melted_beta$Threads <- as.factor(data_melted_beta$Threads)

data_melted_beta$Threads <- ordered(data_melted_beta$Threads , levels = c("8 Threads", "16 Threads", "32 Threads"))

data_melted_beta_memory <- as.data.frame(data_melted_beta %>% filter(Param == "Memory"))
unique(data_melted_beta_memory$Param)

## [1] "Memory"

data_melted_beta_memory$Threads <- ordered(data_melted_beta_memory$Threads , levels = c("8 Threads", "16 Threads", "32 Threads"))

data_melted_beta_memory$Value <- as.numeric(data_melted_beta_memory$Value)
data_melted_beta_memory$Sample <- as.factor(data_melted_beta_memory$Sample)
data_melted_beta_memory$Run <- as.factor(data_melted_beta_memory$Run)

data_melted_beta_memory <- data_melted_beta_memory[order(data_melted_beta_memory$Sample, data_melted_beta_memory$Value),] 

#range(data_melted_beta_memory$Value) #30649908 38445368

data_melted_beta_memory <- data_melted_beta_memory %>% 
  group_by(Sample, Threads) %>% 
  dplyr::mutate(mem_order = sum(Value)) %>% 
  arrange(mem_order)
#class(data_melted_beta_memory) - "grouped_df" "tbl_df"     "tbl"        "data.frame"
data_melted_beta_memory <- as.data.frame(data_melted_beta_memory)


#Memory
# data_melted_beta_memory %>% filter(Sample != "NA12878_Small") %>% 
#   #mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
#   ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run)) + #Dividing by 1000000 to convert kbytes to GB
#   geom_line(aes(color = Run, linetype = Run)) +
#   scale_color_manual(values = c("black", "gray10", "gray50")) +
#   facet_wrap(~Threads) + 
#   labs(y = "Memory (Gigabytes)", x="") + scale_y_continuous() +
#   theme_bw() + theme(legend.title = element_blank()) + #Base, bw, excel_new(), few, light, lindraw
#   theme(strip.background =element_rect(fill="white", colour = "white"))+
#   theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
#   #theme(strip.text.x = element_blank()) +
#   theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) #+
#   #geom_text(x = 6, y = max(data_melted_beta_memory$Value), aes(label = label), 
#     #    data = dat_text, check_overlap = TRUE, inherit.aes = FALSE) 


data_melted_beta_memory %>% 
 # mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) + #note that memory is in kb, dividing by 1000000 to convert it to gb (Maximum resident set size (kbytes))
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  #geom_xspline(size=0.5)  
  scale_color_manual(values = c("firebrick4", "dodgerblue4", "gray50")) +
  facet_wrap(~Threads) + 
  labs(y = "Memory (Gigabytes)", x="") + scale_y_continuous(limits = c(0,40)) + #rescaling to range from 0-40 GB
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))  

# Without re-scaling
data_melted_beta_memory %>% 
 # mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) + #note that memory is in kb, dividing by 1000000 to convert it to gb (Maximum resident set size (kbytes))
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  #geom_xspline(size=0.5)  
  scale_color_manual(values = c("firebrick4", "dodgerblue4", "gray50")) +
  facet_wrap(~Threads) + 
  labs(y = "Memory (Gigabytes)", x="") + scale_y_continuous() + #rescaling to range from 0-40 GB
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))  

# 30-40 scale
data_melted_beta_memory %>% 
 # mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) + #note that memory is in kb, dividing by 1000000 to convert it to gb (Maximum resident set size (kbytes))
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  #geom_xspline(size=0.5)  
  scale_color_manual(values = c("firebrick4", "dodgerblue4", "gray50")) +
  facet_wrap(~Threads) + 
  labs(y = "Memory (Gigabytes)", x="") + scale_y_continuous(limits = c(30,40)) + #rescaling to range from 0-40 GB
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))  

# Applying shorter sample lables:
leg.txt <- levels(data_melted_beta_memory$Sample)
x.labels <- structure(LETTERS[seq_along(leg.txt)], 
                      .Names = leg.txt) #Our sample names are too long - using letters that will be mapped back to the sample names

p1 <- data_melted_beta_memory %>% 
 # mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) + #note that memory is in kb, dividing by 1000000 to convert it to gb (Maximum resident set size (kbytes))
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  #geom_xspline(size=0.5)  
  scale_color_manual(values = c("firebrick4", "dodgerblue4", "gray50")) +
  facet_wrap(~Threads) + 
  labs(y = "Memory (Gigabytes)", x="") + scale_y_continuous(limits = c(30,40)) + #rescaling to range from 0-40 GB
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 0,hjust = 1,vjust = 0.5, size=10)) +scale_x_discrete(name = "", labels = x.labels)

legend <-  get_legend(data_melted_beta_memory %>% 
    mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) +
     geom_line() +
     scale_color_manual(values = c("black", "gray10", "gray50")) +
     facet_wrap(~Threads) + 
    labs(y = "Memory (Gigabytes)", x="") + scale_y_continuous(limits = c(30,40)) +
     theme_bw() + theme(legend.title = element_blank(), legend.position = "bottom") + #Base, bw, excel_new(), few, light, lindraw
     theme(strip.background =element_rect(fill="white", colour = "white"))+
     theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
         geom_point(aes(shape = Sample), alpha = 0) + 
   scale_shape_manual(name = "Samples", values = x.labels) + 
     guides(shape = guide_legend(override.aes = list(size = 2.5, alpha = 1))) + 
     scale_x_discrete(name = "Samples", labels = x.labels))
  
grid.arrange(p1, legend, nrow=2, ncol=1)

# Representation 2
data_melted_beta_memory %>% 
  #mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
    ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run)) +
  geom_point(aes(color=factor(Run),shape=factor(Threads),fill=factor(Threads)), size=2)+
  #geom_line()+
  scale_shape_manual(values=c(3, 17, 16))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  #scale_color_manual(values = c("darkred", "dodgerblue4", "orange")) +
 # facet_wrap(~Threads) + 
  #labs(y = "Mapping Speed (Million of reads/hour", x="") +
  labs(y = "Memory (Gigabytes)", x="")+ scale_y_continuous(limits = c(30,40)) +
  theme_bw() + theme(legend.title = element_blank()) + #Base, bw, excel_new(), few, light, lindraw
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  #theme(strip.text.x = element_blank()) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) #+

  #geom_text(x = 6, y = max(data_melted_beta_memory$Value), aes(label = label), 
    #    data = dat_text, check_overlap = TRUE, inherit.aes = FALSE) 

#Plot 1 - Splitting up plots by thread
# 8 Threads
data_melted_beta_memory %>% filter(Threads == "8 Threads") %>% 
  #mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
    ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) +
  #geom_line(aes(color = Run, linetype = Run)) + #in
    geom_point(color="white") +   geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  #scale_color_manual(values = c("black", "gray40", "gray80")) +
    scale_color_manual(values = c("firebrick4", "dodgerblue4", "gray50")) +
  #facet_wrap(~Threads) + 
  labs(y = paste0("Memory - Gigabytes", "\n", "(8 Threads)"), x="") + scale_y_continuous(limits = c()) +
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

# 16 Threads
data_melted_beta_memory %>% filter(Threads == "16 Threads") %>% 
  #mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
    ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) +
  #geom_line(aes(color = Run, linetype = Run)) +
  geom_point(color="white") +   geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
    scale_color_manual(values = c("firebrick4", "dodgerblue4", "gray50")) +
  #facet_wrap(~Threads) + 
  labs(y = paste0("Memory - Gigabytes", "\n", "(16 Threads)"), x="") + scale_y_continuous(limits = c()) +
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

# 32 threads
data_melted_beta_memory %>% filter(Threads == "32 Threads") %>% 
  #mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
    ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) +
  #geom_line(aes(color = Run, linetype = Run)) +
  geom_point(color="white") +   geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
    scale_color_manual(values = c("firebrick4", "dodgerblue4", "gray50")) +
  labs(y = paste0("Memory - Gigabytes", "\n", "(32 Threads)"), x="") + scale_y_continuous(limits = c()) +
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

#Plot 2
# 8 Threads
data_melted_beta_memory %>% filter(Threads == "8 Threads") %>% 
 # mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
    ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) +
  geom_point(aes(color=factor(Run),shape=factor(Run),fill=factor(Run)), size=2)+
  scale_shape_manual(values=c(3, 17, 16))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Memory - Gigabytes", "\n", "(8 Threads)"), x="") + scale_y_continuous(limits = c()) +
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

# 16 Threads
data_melted_beta_memory %>% filter(Threads == "16 Threads") %>% 
 # mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
    ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) +
  geom_point(aes(color=factor(Run),shape=factor(Run),fill=factor(Run)), size=2)+
  scale_shape_manual(values=c(3, 17, 16))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Memory - Gigabytes", "\n", "(16 Threads)"), x="") + scale_y_continuous(limits = c()) +
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

# 32 threads
data_melted_beta_memory %>% filter(Threads == "32 Threads") %>% 
 # mutate(Sample = fct_reorder2(Sample, Threads, mem_order,  .desc = FALSE)) %>%
    ggplot(aes(x = reorder(Sample, (Value)/1000000), y = (Value)/1000000, group=Run,color=factor(Run))) +
  geom_point(aes(color=factor(Run),shape=factor(Run),fill=factor(Run)), size=2)+
  scale_shape_manual(values=c(3, 17, 16))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Memory - Gigabytes", "\n", "(32 Threads)"), x="") + scale_y_continuous(limits = c()) +
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

#extracting max memory used per run and thread for qsub mem allocation guage

data_melted_beta_memory %>% filter(Run == "STAR" & Threads == "8 Threads") %>% select(Value) %>% max() #31041464

## [1] 31041464

data_melted_beta_memory %>% filter(Run == "STAR" & Threads == "16 Threads") %>% select(Value) %>% max() #32991776

## [1] 32991776

data_melted_beta_memory %>% filter(Run == "STAR" & Threads == "32 Threads") %>% select(Value) %>% max() #36892708

## [1] 36892708

#
data_melted_beta_memory %>% filter(Run == "WASP" & Threads == "8 Threads") %>% select(Value) %>% max() #31041288

## [1] 31041288

data_melted_beta_memory %>% filter(Run == "WASP" & Threads == "16 Threads") %>% select(Value) %>% max() #32991976

## [1] 32991976

data_melted_beta_memory %>% filter(Run == "WASP" & Threads == "32 Threads") %>% select(Value) %>% max() #36892620

## [1] 36892620

data_melted_beta_memory %>% filter(Run == "STAR_WASP" & Threads == "8 Threads") %>% select(Value) %>% max()  #31519628

## [1] 31519628

data_melted_beta_memory %>% filter(Run == "STAR_WASP" & Threads == "16 Threads") %>% select(Value) %>% max()  #33828116

## [1] 33828116

data_melted_beta_memory %>% filter(Run == "STAR_WASP" & Threads == "32 Threads") %>% select(Value) %>% max()  #38445368

## [1] 38445368

3.2 Mapping Speed (Mapping_speed_Million_of_reads_per_hour)

** Not considered as WASP’s speed is based off filtered reads, considering wall clock instead (next section)

data_melted_beta_speed <- data_melted_beta[order(data_melted_beta$Sample, data_melted_beta$Mapping_speed_Million_of_reads_per_hour),] 

data_melted_beta_speed$Threads <- ordered(data_melted_beta_speed$Threads , levels = c("8 Threads", "16 Threads", "32 Threads"))

data_melted_beta_speed$Value <- as.numeric(data_melted_beta_speed$Mapping_speed_Million_of_reads_per_hour)
data_melted_beta_speed$Sample <- as.factor(data_melted_beta_speed$Sample)
data_melted_beta_speed$Run <- as.factor(data_melted_beta_speed$Run)

data_melted_beta_speed <- data_melted_beta_speed[order(data_melted_beta_speed$Sample, data_melted_beta_speed$Value),] 


data_melted_beta_speed <- data_melted_beta_speed %>% 
  group_by(Sample, Threads) %>% 
  dplyr::mutate(speed_order = sum(Mapping_speed_Million_of_reads_per_hour)) %>% 
  arrange(speed_order)

data_melted_beta_speed <- as.data.frame(data_melted_beta_speed)

# data_melted_beta_speed %>% 
#   ggplot(aes(x = Sample, y = Mapping_speed_Million_of_reads_per_hour, group=Run)) +
#   geom_line(aes(color = Run, linetype = Run)) + 
#   scale_color_manual(values = c("darkred", "steelblue", "cyan3")) +
#   facet_wrap(~Threads) + labs(y = "Mapping_Speed", x="") +
#   theme_light() +  theme(legend.title = element_blank()) + #Base, bw, excel_new(), few, light, lindraw
#   #theme(strip.background =element_rect(fill="white"))+
#   # theme(strip.text = element_text(colour = 'black')) +
#   theme(strip.text.x = element_blank()) + theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))# +
# #geom_text(x = 3, y = max(data_melted_beta_memory$Value), aes(label = label), data = dat_text, check_overlap = TRUE) 


data_melted_beta_speed %>% 
  #mutate(Sample = fct_reorder2(Sample, Threads, speed_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, Mapping_speed_Million_of_reads_per_hour), y = Mapping_speed_Million_of_reads_per_hour, group=Run, color=factor(Run))) +
 # geom_line(aes(color = Run, linetype = Run)) + #stat_smooth()
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  #scale_color_manual(values = c("black", "gray50", "gray80" )) + #"darkred", "dodgerblue4", "orange"
    scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  facet_wrap(~Threads) + 
  #labs(y = "Mapping Speed (Million of reads/hour", x="") +
  labs(y = paste0("Mapping Speed", "\n", "(Million reads/hour)"), x = "")+
  theme_bw() + theme(legend.title = element_blank()) + #Base, bw, excel_new(), few, light, lindraw
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  #theme(strip.text.x = element_blank()) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))  #+

  #geom_text(x = 6, y = max(data_melted_beta_memory$Value), aes(label = label), 
    #    data = dat_text, check_overlap = TRUE, inherit.aes = FALSE) 
 
 
data_melted_beta_speed %>% 
  mutate(Sample = fct_reorder2(Sample, Threads, speed_order,  .desc = FALSE)) %>%
  ggplot(aes(x = Sample, y = Mapping_speed_Million_of_reads_per_hour, group=Run)) +
  geom_point(aes(color=factor(Run),shape=factor(Run),fill=factor(Run)), size=2)+
  #geom_line()+
  scale_shape_manual(values=c(3, 17, 16))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  #scale_color_manual(values = c("darkred", "dodgerblue4", "orange")) +
 facet_wrap(~Threads) + 
  #labs(y = "Mapping Speed (Million of reads/hour", x="") +
  labs(y = paste0("Mapping Speed", "\n", "(Million reads/hour)"), x="")+
  theme_bw() + theme(legend.title = element_blank()) + #Base, bw, excel_new(), few, light, lindraw
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  #theme(strip.text.x = element_blank()) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) #+

  #geom_text(x = 6, y = max(data_melted_beta_memory$Value), aes(label = label), 
    #    data = dat_text, check_overlap = TRUE, inherit.aes = FALSE) 


#Plot 1
#8 Threads
data_melted_beta_speed %>% filter (Threads == "8 Threads") %>% 
  #mutate(Sample = fct_reorder2(Sample, Threads, speed_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, Value), y = Mapping_speed_Million_of_reads_per_hour, group=Run, color=Run)) +
  #geom_line(aes(color = Run, linetype = Run)) + 
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Mapping Speed", "\n", "(Million reads/hour)", "\n", "8 Threads"), x = "")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))  

#16 Threads
data_melted_beta_speed %>% filter(Threads == "16 Threads") %>% 
  #mutate(Sample = fct_reorder2(Sample, Threads, speed_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, Value), y = Mapping_speed_Million_of_reads_per_hour, group=Run, color=Run)) +
  #geom_line(aes(color = Run, linetype = Run)) + 
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Mapping Speed", "\n", "(Million reads/hour)", "\n", "16 Threads"), x = "")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))  

#32 Threads
data_melted_beta_speed %>% filter(Threads == "32 Threads") %>% 
 # mutate(Sample = fct_reorder2(Sample, Threads, speed_order,  .desc = FALSE)) %>%
  ggplot(aes(x = reorder(Sample, Value), y = Mapping_speed_Million_of_reads_per_hour, group=Run, color=Run)) +
  #geom_line(aes(color = Run, linetype = Run)) + 
   geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Mapping Speed", "\n", "(Million reads/hour)", "\n", "32 Threads"), x = "")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

3.3 Speed Based on Wall Clock

Note that each sample should have the same number of input reads at the start for each run, however, for WASP, the Log.final.out file reports input reads passed or already filtered reads which are less than the initial input reads (also affects mapping speed, see example below for HG00512). We therefore need to consider speed based on the wall clock, i.e overall time taken for the run to complete

unique(data_melted_beta_speed[,c(1, 4:7)] %>% filter (Sample == "HG00512"))

##      Sample    Threads       Run Number_of_input_reads
## 1   HG00512  8 Threads      WASP               1110628
## 23  HG00512  8 Threads STAR_WASP              66055710
## 45  HG00512  8 Threads      STAR              66055710
## 67  HG00512 16 Threads      WASP               1110628
## 89  HG00512 16 Threads STAR_WASP              66055710
## 111 HG00512 16 Threads      STAR              66055710
## 133 HG00512 32 Threads      WASP               1110628
## 155 HG00512 32 Threads STAR_WASP              66055710
## 177 HG00512 32 Threads      STAR              66055710
##     Mapping_speed_Million_of_reads_per_hour
## 1                                    266.55
## 23                                   581.42
## 45                                   699.41
## 67                                   266.55
## 89                                   925.29
## 111                                 1219.49
## 133                                  266.55
## 155                                 1321.11
## 177                                 1857.82

data_melted_beta_wall_clock <- data_melted_beta %>% filter(Param == "Wall_Clock")

# Extracting number of input reads per sample based on STAR runs - should be same input reads for each sample/fastq files
num_input_reads_per_sample <- unique(data_melted_beta %>% filter(Run == "STAR") %>% select("Sample", "Number_of_input_reads"))
colnames(num_input_reads_per_sample)[2] <- "Number_of_input_reads_initial"
data_melted_beta_wall_clock <- inner_join(data_melted_beta_wall_clock, num_input_reads_per_sample, by = "Sample")

### Number of input reads per sample:
num_input_reads_per_sample %>%
  ggplot(aes(reorder(x=Sample, -Number_of_input_reads_initial), y=Number_of_input_reads_initial)) +
  geom_bar(stat = "identity", fill="gray60") +
  theme_light() +
  scale_y_continuous(expand=c(0,0))+
  scale_x_discrete(expand=c(0,0)) +
  labs(y="Number of Input Reads", x = "") +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))

# Plotting by average input read length
input_read_len_per_sample <- unique(data_melted_beta %>% filter(Run == "STAR") %>% select("Sample", "Average_input_read_length"))
input_read_len_per_sample %>%
  ggplot(aes(reorder(x=Sample, -Average_input_read_length), y=Average_input_read_length)) +
  geom_bar(stat = "identity", fill="gray60") +
  theme_light() +
  scale_y_continuous(expand=c(0,0), limits = c(0, max(input_read_len_per_sample$Average_input_read_length)+10))+
  scale_x_discrete(expand=c(0,0)) +
  labs(y="Average input read length", x = "") +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))

min(num_input_reads_per_sample$Number_of_input_reads_initial)#56254714

## [1] 56254714

max(num_input_reads_per_sample$Number_of_input_reads_initial)#128402941

## [1] 128402941

mean(num_input_reads_per_sample$Number_of_input_reads_initial)#86852506

## [1] 86852506

median(num_input_reads_per_sample$Number_of_input_reads_initial)#92285172

## [1] 92285172

data_melted_beta_wall_clock$Unique_ID <- 1:nrow(data_melted_beta_wall_clock)
data_melted_beta_wall_clock$Unique_ID <- as.character(data_melted_beta_wall_clock$Unique_ID)
data_melted_beta_wall_clock$Wall_Clock <- data_melted_beta_wall_clock$Value #maintaining a copy of the original variable

#Note that we have hours:minutes:seconds and minutes:seconds.milliseconds in our dataset - need to convert all to hours
data_melted_beta_wall_clock_min_sec_ms <- data_melted_beta_wall_clock[grepl('\\.', data_melted_beta_wall_clock$Value), ] #we want to grep all values that have a dot in them = milliseconds - overall these values have minutes:seconds:milliseconds

data_melted_beta_wall_clock_min_sec_ms_split1 <- cSplit(data_melted_beta_wall_clock_min_sec_ms, "Value", ":", direction = "wide", type.convert="as.character")

data_melted_beta_wall_clock_min_sec_ms_split <- cSplit(data_melted_beta_wall_clock_min_sec_ms_split1, "Value_2", ".", direction = "wide", type.convert="as.character")

colnames(data_melted_beta_wall_clock_min_sec_ms_split)[11] <- "minutes"
colnames(data_melted_beta_wall_clock_min_sec_ms_split)[12] <- "seconds"
colnames(data_melted_beta_wall_clock_min_sec_ms_split)[13] <- "milliseconds"


data_melted_beta_wall_clock_min_sec_ms_split$minutes <- as.numeric(as.character(data_melted_beta_wall_clock_min_sec_ms_split$minutes))
data_melted_beta_wall_clock_min_sec_ms_split$seconds <- as.numeric(as.character(data_melted_beta_wall_clock_min_sec_ms_split$seconds))
data_melted_beta_wall_clock_min_sec_ms_split$milliseconds <- as.numeric(as.character(data_melted_beta_wall_clock_min_sec_ms_split$milliseconds))

#Next we shall convert the different units to hours and sum them up for total time in hours
data_melted_beta_wall_clock_min_sec_ms_split <- data_melted_beta_wall_clock_min_sec_ms_split %>%  mutate(min_in_hrs = (minutes/60))
data_melted_beta_wall_clock_min_sec_ms_split <- data_melted_beta_wall_clock_min_sec_ms_split %>%  mutate(seconds_in_hrs = (seconds/(60*60)))
data_melted_beta_wall_clock_min_sec_ms_split <- data_melted_beta_wall_clock_min_sec_ms_split %>%  mutate(milliseconds_in_hrs = (milliseconds/(3.6e+6)))
data_melted_beta_wall_clock_min_sec_ms_split <- data_melted_beta_wall_clock_min_sec_ms_split %>% mutate(time_hrs = (min_in_hrs + seconds_in_hrs + milliseconds_in_hrs))

#Second Subset - overall, these values have hours:minutes:seconds  
data_melted_beta_wall_clock_hr_min_sec  <- as.data.frame(data_melted_beta_wall_clock %>% filter(Unique_ID %nin% data_melted_beta_wall_clock_min_sec_ms$Unique_ID))  #values should not have dots/ms unit values in them
data_melted_beta_wall_clock_hr_min_sec_split <- cSplit(data_melted_beta_wall_clock_hr_min_sec, "Value", ":", direction = "wide", type.convert="as.character")
colnames(data_melted_beta_wall_clock_hr_min_sec_split)[11] <- "hrs"
colnames(data_melted_beta_wall_clock_hr_min_sec_split)[12] <- "minutes"
colnames(data_melted_beta_wall_clock_hr_min_sec_split)[13] <- "seconds"

data_melted_beta_wall_clock_hr_min_sec_split$hrs <- as.numeric(as.character(data_melted_beta_wall_clock_hr_min_sec_split$hrs))
data_melted_beta_wall_clock_hr_min_sec_split$minutes <- as.numeric(as.character(data_melted_beta_wall_clock_hr_min_sec_split$minutes))
data_melted_beta_wall_clock_hr_min_sec_split$seconds <- as.numeric(as.character(data_melted_beta_wall_clock_hr_min_sec_split$seconds))

#Next we shall convert the different units to hours and sum them up for total time in hours
data_melted_beta_wall_clock_hr_min_sec_split <- data_melted_beta_wall_clock_hr_min_sec_split %>%  mutate(min_in_hrs = (minutes/60))
data_melted_beta_wall_clock_hr_min_sec_split <- data_melted_beta_wall_clock_hr_min_sec_split %>%  mutate(seconds_in_hrs = (seconds/(60*60)))
data_melted_beta_wall_clock_hr_min_sec_split <- data_melted_beta_wall_clock_hr_min_sec_split %>% mutate(time_hrs = (hrs + min_in_hrs + seconds_in_hrs))

#Mutating speed based on wall clock (num of input reads/ time in hrs)
data_melted_beta_wall_clock_min_sec_ms_split <- data_melted_beta_wall_clock_min_sec_ms_split %>%  mutate(reads_per_hour = (Number_of_input_reads_initial/time_hrs))
data_melted_beta_wall_clock_hr_min_sec_split <- data_melted_beta_wall_clock_hr_min_sec_split %>%  mutate(reads_per_hour =  (Number_of_input_reads_initial/time_hrs))

head(data_melted_beta_wall_clock_min_sec_ms_split)

##                      Sample      Param    Threads  Run Number_of_input_reads
## 1:                  HG00512 Wall_Clock 16 Threads STAR              66055710
## 2:                  HG00513 Wall_Clock 16 Threads STAR              58601893
## 3:                  HG00731 Wall_Clock 16 Threads STAR              56254714
## 4:                  HG00732 Wall_Clock 16 Threads STAR              70029452
## 5:                  HG00733 Wall_Clock 16 Threads STAR              92075712
## 6: NA12878_Nucleus_nonPolyA Wall_Clock 16 Threads STAR             110469791
##    Mapping_speed_Million_of_reads_per_hour Average_input_read_length
## 1:                                 1219.49                       102
## 2:                                 1205.52                       102
## 3:                                 1205.46                       102
## 4:                                 1183.60                       102
## 5:                                 1205.35                       102
## 6:                                  842.57                       152
##    Number_of_input_reads_initial Unique_ID Wall_Clock minutes seconds
## 1:                      66055710         1    3:26.73       3      26
## 2:                      58601893         2    3:05.82       3       5
## 3:                      56254714         3    2:58.48       2      58
## 4:                      70029452         4    3:43.33       3      43
## 5:                      92075712         5    4:46.23       4      46
## 6:                     110469791         6    8:05.80       8       5
##    milliseconds min_in_hrs seconds_in_hrs milliseconds_in_hrs   time_hrs
## 1:           73 0.05000000    0.007222222        2.027778e-05 0.05724250
## 2:           82 0.05000000    0.001388889        2.277778e-05 0.05141167
## 3:           48 0.03333333    0.016111111        1.333333e-05 0.04945778
## 4:           33 0.05000000    0.011944444        9.166667e-06 0.06195361
## 5:           23 0.06666667    0.012777778        6.388889e-06 0.07945083
## 6:           80 0.13333333    0.001388889        2.222222e-05 0.13474444
##    reads_per_hour
## 1:     1153962703
## 2:     1139855928
## 3:     1137429066
## 4:     1130353029
## 5:     1158901778
## 6:      819846721

head(data_melted_beta_wall_clock_hr_min_sec_split)

##                          Sample      Param    Threads  Run
## 1: NA12878_Nucleus_nonPolyA_Rep Wall_Clock 16 Threads WASP
## 2:        NA12878_Nucleus_PolyA Wall_Clock 16 Threads WASP
## 3:    NA12878_Nucleus_PolyA_Rep Wall_Clock 16 Threads WASP
## 4:                NA12878_PolyA Wall_Clock 16 Threads WASP
## 5:            NA12878_PolyA_Rep Wall_Clock 16 Threads WASP
## 6:                NA12878_Total Wall_Clock 16 Threads WASP
##    Number_of_input_reads Mapping_speed_Million_of_reads_per_hour
## 1:               8737811                                  593.51
## 2:              11671879                                  600.27
## 3:              10382396                                  622.94
## 4:              11045656                                  473.39
## 5:              11009414                                  471.83
## 6:              10550334                                  457.60
##    Average_input_read_length Number_of_input_reads_initial Unique_ID Wall_Clock
## 1:                       152                     106919251       103    1:03:19
## 2:                       152                     128402941       104    1:21:37
## 3:                       152                     116517502       105    1:14:06
## 4:                       202                      97548052       106    1:11:33
## 5:                       202                      93555584       107    1:11:21
## 6:                       202                     105089150       108    1:15:16
##    hrs minutes seconds min_in_hrs seconds_in_hrs time_hrs reads_per_hour
## 1:   1       3      19  0.0500000    0.005277778 1.055278      101318585
## 2:   1      21      37  0.3500000    0.010277778 1.360278       94394647
## 3:   1      14       6  0.2333333    0.001666667 1.235000       94346155
## 4:   1      11      33  0.1833333    0.009166667 1.192500       81801301
## 5:   1      11      21  0.1833333    0.005833333 1.189167       78673231
## 6:   1      15      16  0.2500000    0.004444444 1.254444       83773459

# Subsetting datasets to required plotting variables
data_melted_beta_wall_clock_min_sec_ms_split <- data_melted_beta_wall_clock_min_sec_ms_split[,c(1:5,17:18)]
data_melted_beta_wall_clock_hr_min_sec_split <- data_melted_beta_wall_clock_hr_min_sec_split[,c(1:5,16:17)]

data_melted_beta_wall_clock_converted <- rbind(data_melted_beta_wall_clock_min_sec_ms_split, data_melted_beta_wall_clock_hr_min_sec_split)

# Plotting wall clock
data_melted_beta_wall_clock_converted %>% 
  #mutate(Sample = fct_reorder(Sample, reads_per_hour)) %>%
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run, color=Run)) +
  #geom_line(aes(color = Run, linetype = Run)) +
 geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  scale_color_manual(values = c("black", "gray50", "gray80" )) + 
  facet_wrap(~Threads) + 
  labs(y = "Million Reads/Hour", x = "")+
  theme_bw() + theme(legend.title = element_blank()) + scale_y_continuous(trans='log10') +
  theme(strip.background = element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

  data_melted_beta_wall_clock_converted %>% 
  #mutate(Sample = fct_reorder(Sample, reads_per_hour)) %>%
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run, color=Run)) +
  #geom_line(aes(color = Run, linetype = Run)) +
 geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  facet_wrap(~Threads) + 
  labs(y = "Million Reads/Hour", x = "")+
  theme_bw() + theme(legend.title = element_blank()) + scale_y_continuous(trans='log10') +
  theme(strip.background = element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

# 8 threads
data_melted_beta_wall_clock_converted %>% filter(Threads == "8 Threads") %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run, color=Run)) +
   geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  scale_color_manual(values = c("black", "gray50", "gray80" )) + 
  labs(y = paste0("Million Reads/Hour", "\n", "(8 Threads)"), x = "")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

# 16 threads
data_melted_beta_wall_clock_converted %>% filter(Threads == "16 Threads") %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run, color=Run)) +
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  scale_color_manual(values = c("black", "gray50", "gray80" )) + 
  labs(y = paste0("Million Reads/Hour", "\n", "(16 Threads)"), x = "")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

# 32 threads
data_melted_beta_wall_clock_converted %>%  filter(Threads == "32 Threads") %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run, color=Run)) +
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.5) +
  scale_color_manual(values = c("black", "gray50", "gray80" )) + 
  labs(y = paste0("Million Reads/Hour", "\n", "(32 Threads)"), x = "")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

# Representation 2
data_melted_beta_wall_clock_converted %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run)) +
  geom_point(aes(color=factor(Run),shape=factor(Run),fill=factor(Run)), size=2)+#,alpha=.8
  scale_shape_manual(values=c(3, 17, 16))+
  facet_wrap(~Threads) + 
  #scale_color_manual(values=c('gray50','firebrick3', '#56B4E9'))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = "Million Reads/Hour", x="")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x =  element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

data_melted_beta_wall_clock_converted %>% filter(Threads == "8 Threads") %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run)) +
  geom_point(aes(color=factor(Run),shape=factor(Run),fill=factor(Run)), size=2)+#,alpha=.8
  scale_shape_manual(values=c(3, 17, 16))+
  #scale_color_manual(values=c('gray50','firebrick3', '#56B4E9'))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Million Reads/Hour", "\n", "(8 Threads)"), x="")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x =  element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

data_melted_beta_wall_clock_converted %>% filter(Threads == "16 Threads") %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run)) +
  geom_point(aes(color=factor(Run),shape=factor(Run),fill=factor(Run)), size=2)+#,alpha=.8
  scale_shape_manual(values=c(3, 17, 16))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Million Reads/Hour", "\n", "(16 Threads)"), x="")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x =  element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

data_melted_beta_wall_clock_converted %>% filter(Threads == "32 Threads") %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour/1000000), y = reads_per_hour/1000000, group=Run)) +
  geom_point(aes(color=factor(Run),shape=factor(Run),fill=factor(Run)), size=2)+#,alpha=.8
  scale_shape_manual(values=c(3, 17, 16))+
  scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Million Reads/Hour", "\n", "(32 Threads)"), x="")+
  theme_bw() + theme(legend.title = element_blank()) + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x =  element_markdown(hjust = 0.5, size=12)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

## Passing sample order to number of reads per sample plot 
order_df <- c("NA12878_PolyA", "NA12878_Total", "NA12878_PolyA_Rep", "NA12878_Total_Rep", "NA12878_Nucleus_PolyA", "NA12878_Nucleus_nonPolyA_Rep", "NA12878_Nucleus_PolyA_Rep",
                                  "NA12878_Nucleus_nonPolyA", "NA19240", "NA19239", "HG00732", "HG00513", "NA19238", "HG00512", "HG00733", "HG00731")

# num_input_reads_per_sample %>% 
#     ggplot(aes(x= factor(Sample, levels=order_df), y=Number_of_input_reads_initial)) + 
#   geom_bar(stat = "identity", fill="gray60") + 
#   theme_light() + 
#   scale_y_continuous(expand=c(0,0))+ 
#   scale_x_discrete(expand=c(0,0)) +
#   labs("Number of Input Reads", x = "") +
#   theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 


a <- num_input_reads_per_sample %>% 
#  ggplot(aes(reorder(x=Sample, -Number_of_input_reads_initial), y=Number_of_input_reads_initial)) + 
    ggplot(aes(x= factor(Sample, levels=order_df), y=Number_of_input_reads_initial)) + 
  geom_bar(stat = "identity", fill="#999999") + 
  theme_light() + 
  scale_y_continuous(expand=c(0,0)) +#, limits = c(0, max(num_input_reads_per_sample$Number_of_input_reads_initial)+100))+ 
  scale_x_discrete(expand=c(0,0)) +
  labs(y="Number of Input Reads", x = "") +
  theme(axis.text.x = element_blank(), axis.ticks.x = element_blank())
  

b <- data_melted_beta_wall_clock_converted %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour), y = reads_per_hour, group=Run, color=Run)) +
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.6) +
 # scale_color_manual(values = c("black", "gray50", "gray80" )) + 
    scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Reads/Hours", "\n", "(16 Threads)"), x = "")+
  theme_bw() + theme(legend.title = element_blank(), legend.position = "bottom") + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

plot_grid(a, b, ncol = 1, nrow=2, labels = "AUTO", rel_heights = c(0.5,1),  rel_widths = c(1,1), align = 'hv')

## `geom_smooth()` using method = 'loess' and formula 'y ~ x'

## Warning: Graphs cannot be horizontally aligned unless the axis parameter is set.
## Placing graphs unaligned.

## By average input read length
c <- input_read_len_per_sample %>%
  ggplot(aes(x= factor(Sample, levels=order_df), y=Average_input_read_length)) + 
  geom_bar(stat = "identity", fill="gray60") +
  theme_light() +
  scale_y_continuous(expand=c(0,0), limits = c(0, max(input_read_len_per_sample$Average_input_read_length)+10))+
  scale_x_discrete(expand=c(0,0)) +
  labs(y="Average input read length", x = "") +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))

d <- data_melted_beta_wall_clock_converted %>% 
  ggplot(aes(x = reorder(Sample, reads_per_hour), y = reads_per_hour, group=Run, color=Run)) +
  geom_point(color="white") +
  geom_smooth(se=FALSE, linetype="dashed", size=0.6) +
 # scale_color_manual(values = c("black", "gray50", "gray80" )) + 
    scale_color_manual(values=c('#999999','#E69F00', '#56B4E9'))+
  labs(y = paste0("Reads/Hours", "\n", "(16 Threads)"), x = "")+
  theme_bw() + theme(legend.title = element_blank(), legend.position = "bottom") + 
  theme(strip.background =element_rect(fill="white", colour = "white"))+
  theme(strip.text = element_text(colour = 'black'), strip.text.x = element_markdown(hjust = 0)) +
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) 

plot_grid(c, d, ncol = 1, nrow=2, labels = "AUTO", rel_heights = c(1,1),  rel_widths = c(1,1), align = 'hv')

## `geom_smooth()` using method = 'loess' and formula 'y ~ x'

## Warning: Graphs cannot be horizontally aligned unless the axis parameter is set.
## Placing graphs unaligned.

4.1 Overal Read Distribution (Reads with vs those without tags (did not overlap variants))

num_reads <- as.data.frame(read.table("/home/asiimwe/projects/run_env/alpha_star_wasp_comparison/read_nums.txt"))

colnames(num_reads)

## [1] "V1" "V2"

colnames(num_reads) <- c("Number_of_Reads", "Read_File_Path")
num_reads <- cSplit(num_reads, "Read_File_Path", "/", direction = "wide")
num_reads <- num_reads[,-c(2:8, 10)]
num_reads$Read_File_Path_08 <- as.character(num_reads$Read_File_Path_08)
num_reads$Read_File_Path_10 <- as.character(num_reads$Read_File_Path_10)

colnames(num_reads) <- c("Number_of_Reads", "Sample", "Flag")
num_reads$Flag[num_reads$Flag == "WASP_Reads_Sorted_Unique"] <- "Num_Reads"
num_reads$Flag[num_reads$Flag == "STAR_WASP_vW_Tagged_Reads_Unique"] <- "vW_Tagged_Reads"

#Converting data frame to short format
num_reads_reshaped <- reshape(num_reads, idvar = "Sample", timevar = "Flag", direction = "wide")
colnames(num_reads_reshaped)[2] <- "Num_Reads"
colnames(num_reads_reshaped)[3] <- "vW_Tagged_Reads"

# Mutating frequencies
num_reads_reshaped <- num_reads_reshaped %>% mutate("perc_tagged" = ((vW_Tagged_Reads/Num_Reads)*100))
num_reads_reshaped <- num_reads_reshaped %>% mutate("perc_untagged" = 100 - perc_tagged)

# Plotting distributions
num_reads_reshaped_tagged <- as.data.frame(num_reads_reshaped[, c(1,4)])
num_reads_reshaped_notags <- as.data.frame(num_reads_reshaped[, c(1,5)])
colnames(num_reads_reshaped_tagged)[2] <- "perc"
colnames(num_reads_reshaped_notags)[2] <- "perc"
num_reads_reshaped_tagged$Flag <- "vW_Tagged Reads"
num_reads_reshaped_notags$Flag <- "Reads with no Tag"
  
num_reads_reshaped2 <- rbind(num_reads_reshaped_tagged, num_reads_reshaped_notags)


num_reads_reshaped2$Flag <- ordered(num_reads_reshaped2$Flag , levels = c("vW_Tagged Reads", "Reads with no Tag"))

unique(num_reads_reshaped2$Sample)

##  [1] "HG00731"                      "NA12878_Nucleus_PolyA_Rep"   
##  [3] "HG00513"                      "NA12878_RAMPAGE"             
##  [5] "HG00512"                      "NA12878_Nucleus_nonPolyA_Rep"
##  [7] "NA12878_RAMPAGE_Rep"          "NA12878_Nucleus_nonPolyA"    
##  [9] "NA12878_Nucleus_PolyA"        "NA12878_Total"               
## [11] "NA19238"                      "HG00733"                     
## [13] "NA12878_PolyA"                "HG00732"                     
## [15] "NA12878_PolyA_Rep"            "NA12878_Total_Rep"           
## [17] "NA19239"                      "NA19240"

num_reads_reshaped2 <- num_reads_reshaped2 %>% filter(Sample != "NA12878_RAMPAGE_Rep" & Sample != "NA12878_RAMPAGE")

# PLot
num_reads_reshaped2[order(num_reads_reshaped2$Flag, decreasing = T),] %>% 
ggplot() +
  geom_bar(aes(x = Sample,
               y = perc, 
               group = Sample,
               fill = factor(Flag, levels=c("vW_Tagged Reads","Reads with no Tag" ))), 
           stat = "identity", width = 0.99, alpha=0.5) +
  geom_text(aes(x = Sample,
                y = perc,
                label = paste0(round(perc,1), "%")), size=3,
            position = position_stack(vjust = 0.5)) + theme_test() +
 scale_color_manual(values=c('gray40', "gray80"))+ #'#b0b4b6', "#dadedf"   '#f5b85a', "#8baecf"
  #scale_color_manual(values=c('darkorange2', "dodgerblue4"))+
 scale_fill_manual(values=c('gray40', "gray80"))+ #'#b0b4b6', "#dadedf"   '#f5b85a', "#8baecf"
  labs(y="Read Distribution (%)", x="")+
theme(legend.title = element_blank()) +   
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))+
  scale_y_discrete(expand = c(0,0)) + scale_x_discrete(expand = c(0,0))

#ggsave("~/Desktop/dist.pdf", plot=plot1,width=10, height=6, device="pdf")

#Representation 2 - plotting only vW_Tagged read percentages
theme_asiimwe <- function(base_size = 12) {
  theme_bw(base_size = base_size) %+replace%
    theme(
      plot.title = element_text(size = rel(1), face = "bold", margin = margin(0,0,5,0), hjust = 0),
      panel.grid.minor = element_blank(),
      panel.border = element_blank(),
      axis.title = element_text(size = rel(0.85), face = "bold"),
      axis.text = element_text(size = rel(0.70), face = "bold"),
      axis.line = element_line(color = "black", arrow = arrow(length = unit(0.3, "lines"), type = "closed")),
      legend.title = element_text(size = rel(0.85), face = "bold"),
      legend.text = element_text(size = rel(0.70), face = "bold"),
      legend.key = element_rect(fill = "transparent", colour = NA),
      legend.key.size = unit(1.5, "lines"),
      legend.background = element_rect(fill = "transparent", colour = NA),
      strip.background = element_rect(fill = "#17252D", color = "#17252D"),
      strip.text = element_text(size = rel(0.85), face = "bold", color = "white", margin = margin(5,0,5,0))
    )
}

num_reads_reshaped2[order(num_reads_reshaped2$Flag, decreasing = T),] %>%  filter(Flag == "vW_Tagged Reads")  %>%
ggplot() +
  geom_bar(aes(x = reorder(Sample, -perc),
               y = perc, 
               group = Sample,
               fill = factor(Flag, levels=c("vW_Tagged Reads","Reads with no Tag" ))), 
           stat = "identity", width = 0.99, alpha=0.4) +
  geom_text(aes(x = Sample,
                y = perc,
                label = paste0(round(perc,1), "%")), size=3,
            position = position_stack(vjust = 0.5)) + theme_cowplot() +
  scale_color_manual(values=c('darkorange2'))+
    scale_fill_manual(values=c('darkorange2'))+
  labs(y="vW_Tagged Reads (%)", x="")+
theme(legend.title = element_blank(), legend.position = "none") +   
  theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10))+
 scale_y_continuous(expand = c(0,0), limits = c(0, 15)) + scale_x_discrete(expand = c(0,0))

#ggsave("~/Desktop/dist.pdf", plot=plot1,width=10, height=6, device="pdf")

colnames(num_reads_reshaped) <- c("Sample", "Num_Reads",       "vW_Tagged_Reads" ,"Percentage_Tagged" ,    "Percentage_Untagged" )
#Table with percent distribution of reads with and without tags
knitr::kable(
  num_reads_reshaped, row.names = FALSE, #row.names = TRUE, 
 # caption = '<b>Reads</b>', format = 'html'
)

4.2 Sample-specific vW_Tag Distributions for alignments that did not pass WASP filtering

samples <- c("HG00512","HG00513","HG00731","HG00732","HG00733","NA12878_Nucleus_nonPolyA","NA12878_Nucleus_nonPolyA_Rep",
"NA12878_Nucleus_PolyA", "NA12878_Nucleus_PolyA_Rep", "NA12878_PolyA","NA12878_PolyA_Rep","NA12878_Total","NA12878_Total_Rep","NA19238",                     
"NA19239","NA19240")

files_list <- list()
for(i in samples){
path <- paste0("/home/asiimwe/Desktop/TD/pass2_to_keep_", i , ".txt")
files_list[[i]] <- path
}
files = do.call(rbind, files_list)

# Reading and merging data from all samples
merger <- lapply(files, function(x) {
  merger <- read.table(x, comment.char = "", sep = '', header = FALSE, stringsAsFactors = FALSE)
  merger$source <- x
  return(merger)
}) %>%
bind_rows()

merger <- as.data.frame(merger)
merger %>% head(n=2)
merger$Sample <- merger$source
merger$source <- NULL
merger$Sample <- gsub("/home/asiimwe/Desktop/TD/pass2_to_keep_", "", merger$Sample)
merger$Sample <- gsub(".txt", "", merger$Sample)
merger %>% head(n=2)
unique(merger$Sample)#validate called samples

merger <- as.data.frame(merger)
colnames(merger) <- c("Read_ID", "vW_Tag", "Reads_to_remap", "Reads_to_keep", "Sample")
merger %>% head(n=2)

merger <- cSplit(merger, "vW_Tag", ":", direction = "wide", type.convert="as.character")
merger$vW_Tag_1 <- NULL
merger$vW_Tag_2 <- NULL
colnames(merger)[5] <- "vW_Tag"
merger %>% head(n=2)

merger$vW_Tag <- as.integer(as.character(merger$vW_Tag))
merger <- as.data.frame(merger)
write.csv(merger, file="/home/asiimwe/projects/run_env/alpha_star_wasp_comparison/Downstream_Analysis/merger_reads_to_keep_and_remap.csv")

#Creating dataframe with mutated percentages per tag for plotting
perclist <- list()

for (i in unique(merger$Sample)){
  #print(i)
  perc <- merger %>% filter(Sample == i)
  perc <- as.data.frame(perc %>% 
  group_by(vW_Tag) %>%
  dplyr::summarise(cnt = n()) %>%
  mutate(freq = formattable::percent(cnt / sum(cnt))))

  perc$Sample <- i
  perclist[[i]] <- perc
}

vW_Tag_percentages = do.call(rbind, perclist)

merger_vW_Tag_percentages <- as.data.frame(vW_Tag_percentages)
orders <- as.data.frame(merger_vW_Tag_percentages %>% filter(vW_Tag != 1) %>% 
                          group_by(Sample) %>%  
                          dplyr::summarise(sum(cnt)))

colnames(orders)[2] <- "order"
vW_Tag_percentages <- inner_join(merger_vW_Tag_percentages, orders, by = "Sample")

# Plotting
unique(vW_Tag_percentages$vW_Tag) #1 3 4 6 7
colourCount = length(unique(vW_Tag_percentages$vW_Tag))
getPalette = colorRampPalette(brewer.pal(8,"RdYlBu"))#RdYlBu Spectral Greys

vW_Tag_percentages$vW_Tag <- as.factor(vW_Tag_percentages$vW_Tag)

vW_Tag_percentages$Freq <- vW_Tag_percentages$freq
vW_Tag_percentages$Freq <- gsub("%", "", vW_Tag_percentages$Freq)
vW_Tag_percentages$Freq <- as.numeric(as.character(vW_Tag_percentages$Freq))

#Sneak pick into distributions of tags overall
vW_Tag_percentages %>% group_by(vW_Tag) %>% dplyr::summarise(sum(cnt))
# A tibble: 5 × 2
#  vW_Tag `sum(cnt)`
#  <fct>       <int>
#1 1        97368800
#2 3           40555
#3 4          864141
#4 6          571383
#5 7           34311

#https://htmlcolorcodes.com/
#"#D73027" - red var
#"#FA9D59" -orange var
#"#EFE9C4" - light or
#"#9DCDE3" -blue
#"#4575B4" - dark blue

vW_Tag_percentages$vW_Tag_desc <- vW_Tag_percentages$vW_Tag
vW_Tag_percentages$vW_Tag_desc <- as.character(vW_Tag_percentages$vW_Tag_desc)
vW_Tag_percentages$vW_Tag_desc[vW_Tag_percentages$vW_Tag_desc == "1"] <- "Alignment Passed WASP Filtering"
vW_Tag_percentages$vW_Tag_desc[vW_Tag_percentages$vW_Tag_desc == "3"] <- "Variant Base in Read is N/non-ACGT"
vW_Tag_percentages$vW_Tag_desc[vW_Tag_percentages$vW_Tag_desc == "4"] <- "Remapped Read did not Map"
vW_Tag_percentages$vW_Tag_desc[vW_Tag_percentages$vW_Tag_desc == "6"] <- "Remapped Read Maps to Different Locus"
vW_Tag_percentages$vW_Tag_desc[vW_Tag_percentages$vW_Tag_desc == "7"] <- "Read Overlaps too Many Variants"

write.csv(vW_Tag_percentages, file="/home/asiimwe/projects/run_env/alpha_star_wasp_comparison/Downstream_Analysis/vW_Tag_percentages.csv")

vW_Tag_percentages <- read.csv(file="/home/asiimwe/projects/run_env/alpha_star_wasp_comparison/Downstream_Analysis/vW_Tag_percentages.csv")

(plot <- vW_Tag_percentages %>% filter(vW_Tag != 1) %>% 
    #mutate(Sample = fct_reorder(Sample, order)) %>%
    ggplot() +
    geom_bar(aes(x = reorder(Sample,   
                 -Freq), Freq,
                 group = Sample,
                 fill = vW_Tag_desc), 
             stat = "identity", width = 0.99, alpha=0.8) +
    geom_text(aes(x = Sample,
                  y =Freq,
                  label = freq), size = 3, 
              position = position_stack(vjust = 0.5)
              ) +
    #scale_fill_manual(values = getPalette(colourCount))+
    #scale_fill_manual(values = c("#9DCDE3", "#D73027", "#4575B4", "#FA9D59", "#EFE9C4")) +
    scale_fill_manual(values =c("Variant Base in Read is N/non-ACGT" = "#D73027",
                                "Remapped Read did not Map" = "#FA9D59",
                                "Remapped Read Maps to Different Locus" ="#EFE9C4",
                                "Read Overlaps too Many Variants"="#9DCDE3"))+
    theme_test()+
    labs(y=paste0("% Distribution of vW_Tagged", "\n", "Reads that did not pass WASP filtering"), x="") +
    theme(legend.title = element_blank(), legend.position="right")  + #legend.position = c(0.8, 0.65)
    theme(axis.text.x = element_text(angle = 90,hjust = 1,vjust = 0.5, size=10)) +
    scale_y_continuous(expand = c(0.02,0), limits = c(0, 3)) + 
    scale_x_discrete(expand = c(0,0)))# + coord_flip()

#ggsave("~/Desktop/merger/plot_nocolor.pdf", plot=plot,width=10, height=15, device="pdf")
#ggsave("~/Desktop/merger/plot_color.pdf", plot=plot,width=10, height=15, device="pdf")#output with color too

#Flipped
(plot <- vW_Tag_percentages %>% filter(vW_Tag != 1) %>% 
    ggplot() +
    geom_bar(aes(x = reorder(Sample,   
                 Freq), Freq,
                 group = Sample,
                 fill = vW_Tag_desc), 
             stat = "identity", width = 0.99, alpha = 0.7) +
   geom_text(aes(x = Sample,
                  y =Freq,
                  label = freq), size = 3, 
              position = position_stack(vjust = 0.5)
              ) +
scale_fill_manual(values =c("Variant Base in Read is N/non-ACGT" = "#D73027",
                                "Remapped Read did not Map" = "#FA9D59",
                                "Remapped Read Maps to Different Locus" ="#EFE9C4",
                                "Read Overlaps too Many Variants"="#9DCDE3"))+
    theme_test()+
    labs(y=paste0("% Distribution of vW_Tags", "\n", "reads that did not pass WASP filtering"), x="") +
    theme(legend.title = element_blank(), legend.position = "right")  + #, legend.position = "bottom"
    theme(axis.text.x = element_text(angle = 0,hjust = 1,vjust = 0.5, size=10)) +
    scale_y_continuous(expand = c(0.02,0), limits = c(0, 3)) + 
    scale_x_discrete(expand = c(0,0))) + coord_flip()

Venn Diagrams per Sample

merger$to_keep_recoded <- merger$ Reads_to_keep
merger$to_remap_recoded <- merger$Reads_to_remap

sum(is.na(merger$to_remap))#161
sum(is.na(merger$to_keep))#49283

merger$to_remap_recoded <- as.character(merger$to_remap_recoded)
merger$to_keep_recoded <- as.character(merger$to_keep_recoded)

merger$to_remap_recoded[is.na(merger$to_remap_recoded)] <- 0
merger$to_remap_recoded[merger$to_remap_recoded != 0] <- 1
merger$to_keep_recoded[is.na(merger$to_keep_recoded)] <- 0
merger$to_keep_recoded[merger$to_keep_recoded != 0] <- 1
head(merger)

merger$vW_tag <- as.integer(merger$vW_Tag)
merger$to_remap_recoded <- as.integer(merger$to_remap_recoded)
merger$to_keep_recoded <- as.integer(merger$to_keep_recoded)

#Creating comparative read sets for venn diagram
merger$Reads_to_remap <- as.character(merger$Reads_to_remap)
merger$Reads_to_keep <- as.character(merger$Reads_to_keep)

merger$Reads_to_remap[is.na(merger$Reads_to_remap)] <- "0" #Venn diagrams don't work well with NAs so we shall replace all with zeros
merger$Reads_to_keep[is.na(merger$Reads_to_keep)] <- "0"
str(merger)

merger_HG00512 <- merger %>% filter(Sample == "HG00512")
set1 <- as.character(merger_HG00512$Read_ID)
set2 <-  as.character(merger_HG00512$Reads_to_remap)
set3 <-  as.character(merger_HG00512$Reads_to_keep)

myCol <- brewer.pal(3, "Pastel2")


venn.diagram(
  x = list(set1, set2, set3),
  category.names = c("vW_tagged_reads" , "Reads_to_remap" , "Reads_to_keep"),
  filename = '/home/asiimwe/projects/run_env/comparison_star_wasp/genome-wide/venn2.png',
  output = TRUE ,
  imagetype="png" ,
  height = 600 , 
  width = 600 , 
  resolution = 300,
  compression = "lzw",
  lwd = 0.25,
  col=c('#21908dff', '#ffa533', "#2a5c92"),
  fill = c(alpha("#21908dff",0.2), alpha('#ffa533',0.2), alpha('#2a5c92',0.2)),
  cex = 0.5,
  fontfamily = "sans",
  cat.cex = 0.4,
  cat.default.pos = "outer",
  cat.pos = c(-27, 27, 135),
  cat.dist = c(0.055, 0.055, 0.085),
  cat.fontfamily = "sans",
  #cat.col = c("#21908dff", '#ffa533', '#ffa533'),
  cat.col = "black",
  rotation = 1
)