Medical Bioinformatics and e-Bioscience

---

Back to menu

Statistical Computing in R

This course is part of the PhD course program of the AMC Graduate School.

• Location: Academic Medical Center, Amsterdam.
• Course material: All course material is provided during the course.
• ECTS: 0.4
• Teachers: Perry Moerland (coordinator) and Ronald Geskus

Goal

R is a simple programming language for statistical computing. Due to its flexibility and the large variety of statistical functions available in R. it is a popular alternative for programs like SPSS. However, for a beginner the learning curve of mastering R can be rather steep. This course helps the student to become familiar with the basics of R. After the course the student will be able to write short programs in R for basic statistical analyses and for plotting publication-quality figures.

Schedule and course material

Location	Day	Time	Topic	Lectures	Exercises	Answers
L01-243/245	October 3, 2011	09.30-12.30		Notes Slides Example code	pdf	pdf R code
	October 3, 2011	12.30-13.30	Lunch break
L01-243/245	October 3, 2011	13.30-16.30
L01-243/245	October 4, 2011	09.30-12.30
	October 3, 2011	12.30-13.30	Lunch break
L01-243/245	October 4, 2011	13.30-16.30	Biostatistics track
L01-243/245	October 4, 2011	13.30-16.30	Bioinformatics track

Datasets for the exercises

• titanic3.dta
• titanic3select.txt

Information on R

• R homepage
• R and its packages can be downloaded from CRAN
• Introductory material
  • A basic introduction to R, somewhat in line with this course. See also here.
  • This page shows the commands to be used for many aspects of a statistical analysis, and has been created especially for experienced users of some other statistical packages.
  • Other sites to get you started are http://www.burns-stat.com/pages/Tutor/hints_R_begin.html and http://zoonek2.free.fr/UNIX/48_R/all.html.
  • There is also an R wiki with information for starters and much more information
  • There is a book called "R for SAS and SPSS users''. The website for the book is http://rforsasandspssusers.com. This web site also has a freely downloadable early version of the book. It also has a comparison table that lists for many SAS and SPSS procedures the corresponding R packages. See also http://r4stats.com.
• Manuals (basic)
  • An introduction to R (pdf,html): R in 100 pages
  • R reference card: R in 4 pages
• Manuals (advanced)
  • R Data Import/Export (pdf,html)
  • R Language Definition (pdf,html)
  • Writing R Extensions (pdf,html)
  • R Installation and Administration (pdf,html)
• Manuals (graphics)
  • A preprint of the book ggplot2: elegant graphics for data analysis

Pointers

• Editors
  • RStudio provides an intuitive interface to R and powerful coding tools
  • Tinn-R
  • WinEdt together with the RWinEdt plug-in
  • Notepad++ together with the plug-in
  • Relax
• GUIs
  • http://www.sciviews.org/SciViews-K/index.html
  • http://rkward.sourceforge.net
  • http://www-sre.wu-wien.ac.at/SimpleR
  • http://stats.math.uni-augsburg.de/JGR
• Many books have been written that explain how to use R for statistical analysis. See
  • http://www.r-project.org/doc/bib/R-books.html
  • http://www.r-project.org/doc/bib/R-publications.html
  • http://www.math.smith.edu/sasr
• Miscellaneous
  • A web site with many beautiful graphs and the corresponding R code is
  • There exist programs that allow to run R from Excel and from SPSS (search on 'R plug-in' at http://www.spss.com/devcentral)

Information on Bioconductor

Bioconductor is an open source and open development software project for the analysis and comprehension of genomic data in R. Much of data analysis in bioinformatics is done within the R/Bioconductor statistical environment. For example, many statistical methods for the analysis of microarray and other high-throughput data are available from Bioconductor.

• Bioconductor mailing list
• Recommended literature
  • Bioconductor Case Studies, Hahne, F.; Huber, W.; Gentleman, R.; Falcon, S., Spinger, 2008.
  • Bioinformatics and Computational Biology Solutions Using R and Bioconductor, Gentleman, R.; Carey, V.; Huber, W.; Irizarry, R.; Dudoit, S. (Eds.), Springer, 2005.

More information for the bioinformatics track

• The ClassDiscovery package is not on CRAN or Bioconductor but at MDAnderson. Just issue the oompaLite() command mentioned there.

-- PerryMoerland - 2011-09-24