Base R Equivalents of the "Teach the Tidyverse to Beginners" Example

base-code.R

url <- "http://varianceexplained.org/files/Brauer2008_DataSet1.tds"

# Clean and tidy the data
d1 <- rio::import(url, format = "tsv")
d2 <- cbind(d1, setNames(do.call(rbind.data.frame,strsplit(d1$NAME, " ?\\|\\| ?"))[,-5], 
                         c("name", "BP", "MF", "systematic_name")))
d3 <- 
subset(
  within(
    reshape(d2, 
      varying = list(names(d2)[grepl("^.0", names(d2))]),
      v.names = "expression",
      times = names(d2)[grepl("^.0", names(d2))],
      timevar = "sample",
      direction = "long"
    ), {
      nutrient = substring(sample, 1, 1);
      rate = as.numeric(gsub("^.{1}", "", sample))
  }), 
  !is.na(expression) & systematic_name != "", 
  -c(sample,id,NAME,GID,YORF,GWEIGHT)
)

# Visualize a set of four genes
ggplot(subset(d3, BP == "leucine biosynthesis"), 
       aes(rate, expression, color = nutrient)) +
  geom_point() +
  geom_smooth(method = "lm", se = FALSE) +
  facet_wrap(~name + systematic_name)