Here I am just providing some links to various R resources that may be helpful as you start to learn to program and perform analyses in R. A little google-fu and you can find many many more.
Not surprisingly, R studio has some to check out
There are many of these, but here are a few...
DataCamp Tutorials has nice very interactive introductory tutorials for R. In particular the introduction to R, intermediate R, and functions in R "classes" will get you a long way. It has similarly nice introductory Python tutorials. I have set up a class page for Bio720 (2016) so you can use the tutorials.
(less about understanding and performing analyses, more about programming and using R). There are many many more, but these are just a few I have used and read.
Check back on this page for a few suggestions of introductory books.
R for data science. Garrett Grolemund and Hadley Wickham. This is a nice (and free online) book to introduce basic R programming for data analysis within the context of the tidyverse. An excellent introduction to importing, munging and plotting data the tidyverse way.
R programming for bioinformatics. Robert Gentleman. A very nice introduction to programming in R. Not much to do about bioinformatics per se, but a good way of learning R with bioinformatics examples, and some introduction to some of the classes particular to bioconductor.
The R Book. A really nice introduction to using R, data structures in R, plotting and basic statistics. If you are going to be using R, you want to have a copy of this to use!
The art of R programming. A really nice introduction to programming in R. Does not assume you have any background in programming.
R in a nutshell. Once you have some basic understanding of programming in general, this is a nice place to look things up quickly and does a good job of explaining what R is doing with respect to data structures.
R syntax style guide according to google
R syntax style guide according to Hadley Wickham (and thus R-studio)
R syntax style guide according to me
CHECK OUT CRAN task views to see links to libraries with specialized functions.
are available through bioconductor. There are hundreds of amazing libraries. SUper useful tutorials online and example workflows will get you going quickly. One thing to know is that some of the way that data is stored for some bioconductor objects may seem a bit different than other R objects. In particular the two main data structures in bioconductor are the ExpressionSet and SummarizedExperiments. It takes a bit to get used to it, but they can be very helpful down the road!
The Hadley Wickham-verse (also known as the tidyverse). Hadley Wickham has developed some very important and widely used libraries in R, and is one of the chief scientific officers for R-studio. In particular one of his goals was to make doing routine things (data munging, aggregating, cleaning and plotting) in R simpler, more intuitive and more consistent than is found with standard R syntax, and as such many of his libraries are widely used (https://github.com/hadley). Much of the usage of this is clearly demonstrated in the R for data science book.
It is worth noting that of his (HW) effort to do so, he has created a sort of macro language in R (more like a parallel syntax really) that can be quite different. However, while it can be quite different than base R syntax, the learning curve is far less steep, and far more consistent in usage (part of the point). These packages include: dplyr, reshape2, tidyr, readr, devtools, stringr, ggplot2, lubridate, broom. All of these are available from his github page. He has also written several books that cover the gamut of library specific things (ggplot2) to advanced R programming and writing your own packages.
-
While the links above should get you started, everything is also at the github page here. There also seems to be some other libraries that seem to link closely to the tidyverse philosophy like
tidytextfor text mining. data.table has some amazingly useful functions to import and work with large data. -
One thing to note is that the tidyverse and bioconductor represent two fairly distinct ecosystems in
R, see this blog post for some points of comparison (and how to create some tidyverse ideas for bioconductor). Also check out this list of other suggested packages (some are the same). There is also thebiobroomlibrary for "tidying" bioconductor objects.
libraries like data.table provides a very easy to use approach to reading in large data sets (common in genomics) very quickly. readr uses a different approach but with similar goals.
In addition to great tools for analysis and plotting, there are great ways of integrating your code and plots into reports. You can do it the standard LaTeX way in R using Sweave() and Stangle (both part of the base distribution) to integrate with LaTeX (and make PDFs etc). If you want to use other formats (such as the widely used markdown format I use) then the knitr library is essential.
There are a number of tutorials available on writing R packages!
R Packages. Hadley Wickham. A good introduction to writing your own R packages.
Roxygen, Shiny,...
If you are using R-studio, check out https://support.rstudio.com/hc/en-us/sections/200107586-Using-RStudio
for many helpful suggestions.
Not surprisingly, Hadley Wickham (and others) have written several important books, including:
-
Advanced R focused on really understanding how R works and best programming practices.
-
R packages on how to write and build
Rpackages. -
Some may get a kick out of John Chambers book "Software for Data Analysis: Programming with R" on the evolution of
SandR, and how to program in it. -
Rcpp for a great way of integrating c++ code and calling it from
R -
devtools is helpful for developing R libraries
-
testthat for units tests
-
Efficient R Programming by Colin Gillespie and Robin Lovelace. Looks like a great book.
-
Data Visualization: A practical introduction, by Kieran Healy
-
Data visualization. Claus Wilke. This book is as much about good practices on data visualization, along with the
Rcode (ggplot2) used to generate the visualizations.