In insightsengineering/hermes: Preprocessing, analyzing, and reporting of RNA-seq data

Objective

Update and replace the draw_countdist() function from rnaseqTools to use ggplot2 and use an HermesData object as input.

Idea

In order to plot the library depth, we need to access the counts data in the HermesData object.

#Creating the HermesData object from the example SummarizedExperiment object.
result <- hermes_data

#Accessing the counts data.
counts(result)

#Transforming counts data into Log2.
log2(counts(result) + 1)

#Coercing the data into a data frame because ggplot requires a data frame object.
df <- data.frame(log2(counts(result) + 1))

#Tranforming the data into long format.
df.long <- tidyr::gather(df)

#Putting the final function together.
draw_libsize_densities <- function(object, log){
  assert_that(
    isClass(object, "HermesData"),
    is.logical(log)
  )
  if(isTRUE(log)){
    df <- as.data.frame(log2(counts(object) + 1))
  } else {
    df <- as.data.frame(counts(object))
  }
  df.long <- gather(df)
  ggplot(df.long, aes(value, group = key)) +
    geom_density() +
    expand_limits(x = -2.5) +
    ggtitle(ifelse(isTRUE(log), "Log2 Count Distribution", "Count Distribution")) +
    xlab(ifelse(isTRUE(log), "Log2(Count + 1)", "Counts")) +
    ylab("Density")
}

#Running the function.
draw_libsize_densities(result, TRUE)

insightsengineering/hermes documentation built on Dec. 15, 2024, 8:07 a.m.