comparison cpm_tpm_rpk.R @ 1:b74bab5157c4 draft

planemo upload for repository https://github.com/ARTbio/tools-artbio/tree/master/tools/cpm_tpm_rpk commit d46436d5d73356c8803d6d97a110a2754e8a03fb

0:35d032c46a4e

if (length(commandArgs(TRUE)) == 0) {
  system("Rscript cpm_tpm_rpk.R -h", intern = F)
  q("no")
}

# Load necessary packages (install them if it's not the case)
requiredPackages = c('optparse')
for (p in requiredPackages) {
  if (!require(p, character.only = TRUE, quietly = T)) {
    install.packages(p)
  }
  suppressPackageStartupMessages(suppressMessages(library(p, character.only = TRUE)))
}


#Arguments
option_list = list(
  make_option(

  make_option(
    c("-o", "--out"),
    default = "res.tab",
    type = 'character',
    help = "Output name [default : '%default' ]"
  )
)


opt = parse_args(OptionParser(option_list = option_list),
                 args = commandArgs(trailingOnly = TRUE))

if (opt$data == "" & !(opt$help)) {

  return(TPM)
}

data = read.table(
  opt$data,
  h = opt$colnames,
  row.names = 1,
  sep = opt$sep
)

if (opt$type == "tpm" | opt$type == "rpk") {

  row.names = F,
  quote = F,
  sep = "\t"
)

1:b74bab5157c4

if (length(commandArgs(TRUE)) == 0) {
  system("Rscript cpm_tpm_rpk.R -h", intern = F)
  q("no")
}


# load packages that are provided in the conda env
options( show.error.messages=F,
       error = function () { cat( geterrmessage(), file=stderr() ); q( "no", 1, F ) } )
loc <- Sys.setlocale("LC_MESSAGES", "en_US.UTF-8")
warnings()
library(optparse)
library(ggplot2)
library(reshape2)
library(Rtsne)



#Arguments
option_list = list(
  make_option(

  make_option(
    c("-o", "--out"),
    default = "res.tab",
    type = 'character',
    help = "Output name [default : '%default' ]"
  ),
  make_option(
    "--tsne",
    default = FALSE,
    type = 'logical',
    help = "performs T-SNE [default : '%default' ]"
  ),
  make_option(
    "--seed",
    default = 42,
    type = 'integer',
    help = "Seed value for reproducibility [default : '%default' ]"
  ),
  make_option(
    "--perp",
    default = 5.0,
    type = 'numeric',
    help = "perplexity [default : '%default' ]"
  ),
  make_option(
    "--theta",
    default = 1.0,
    type = 'numeric',
    help = "theta [default : '%default' ]"
  ),
  make_option(
    c("-D", "--tsne_out"),
    default = "tsne.pdf",
    type = 'character',
    help = "T-SNE pdf [default : '%default' ]"
  )
)

opt = parse_args(OptionParser(option_list = option_list),
                 args = commandArgs(trailingOnly = TRUE))

if (opt$data == "" & !(opt$help)) {

  return(TPM)
}

data = read.table(
  opt$data,
  header = opt$colnames,
  row.names = 1,
  sep = opt$sep
)

if (opt$type == "tpm" | opt$type == "rpk") {

  row.names = F,
  quote = F,
  sep = "\t"
)

## 
if (opt$tsne == TRUE) {
  df = cpm(data)
  # filter and transpose df for tsne
  df = df[rowSums(df) != 0,] # remove lines without information (with only 0 counts)
  tdf = t(df)
  tdf =  log2(tdf + 1)
  # make tsne and plot results
  set.seed(opt$seed) ## Sets seed for reproducibility
  # Run TSNE
  tsne_out <- Rtsne(tdf, perplexity=opt$perp, theta=opt$theta) # 
  embedding <- as.data.frame(tsne_out$Y)
  embedding$Class <- as.factor(sub("Class_", "", rownames(tdf)))
  gg_legend = theme(legend.position="none")
  ggplot(embedding, aes(x=V1, y=V2)) +
    geom_point(size=1.25, color='red') +
    geom_text(aes(label=Class),hjust=-0.2, vjust=-0.5, size=2.5, color='darkblue') +
    gg_legend +
    xlab("") +
    ylab("") +
    ggtitle('t-SNE of data (log2CPM transformed)')
  ggsave(file=opt$tsne_out, device="pdf")
}