Skip to Content

Rnw to Rmd waste another afternoon


When I read the docmentation of SeqGSEA, I get furious at the prompt (>):

> rcounts <- cbind(t(sapply(1:10, function(x) {rnbinom(5, size=10, prob=runif(1))} )),
+                  t(sapply(1:10, function(x) {rnbinom(5, size=10, prob=runif(1))} )))
> colnames(rcounts) <- c(paste("S", 1:5, sep=""), paste("C", 1:5, sep=""))
> geneIDs <- c(rep("G1", 4), rep("G2", 6))
> exonIDs <- c(paste("E", 1:4, sep=""), paste("E", 1:6, sep=""))
> RCS <- newReadCountSet(rcounts, exonIDs, geneIDs)

So I decided to convert the .Rnw source file into .Rmd.


First, someone kindly recommend me to use this workflow: .Rnw -> .tex -> .md.

After results='hide' and highlight=TRUE, most content seems fine, but there are still a few serious problems.


If you see \newcommand{\Rpackage}[1]{{\textit{#1}}}, then \Rpackage{SeqGSEA} should be **SeqGSEA**. However, it’s very hard to convert every custom command perfectly.

To make things worse, all chunk options were dropped, and it’s very hard to recover. For example, you need to construct fig.align='center', out.width='95%', fig.cap='Gene scores ... data sets.' from the following output:

\caption{Gene scores ... data sets.}


I started the work at about 14:30pm and wrote the post at about 17:12pm. In the interval, I almost rewrote the whole document.

In the process, I realized that parsing output is really wasting time, such as converting .pdf to .docx by Foxit PDF Editor. That might be the reason why the author of knitr hasn’t write a Rnw2Rmd() function.