R software by Henrik Bengtsson

Henrik Bengtsson, 2007-08-21

Packages

Below is a list of all R packages that I am an author or co-author of. For requirements and package dependencies, please see below.

Name	Description	Notes	Status
affxparser	Affymetrix File Parsing SDK. The affxparser package provides methods for fast and memory efficient parsing of Affymetrix files using the Affymetrix' Fusion SDK. Both traditional ASCII- and binary-based files are supported, as well as Affymetrix future binary format "Calvin". The efficiency of the parsing is dependent on whether a specific file is binary or ASCII. This repository provides the developers version of the package. http://www.braju.com/R/affxparser/	Bioconductor; native code	beta & stable
aroma	An R Object-oriented Microarray Analysis package. http://www.maths.lth.se/help/R/aroma/	100% R	alpha, beta & stable
aroma.affymetrix	Basic classes for general Affymetrix data. For more details, see aroma.affymetrix.	100% R	beta
aroma.apd	The Affymetrix Probe-level Data (APD) file format. The (in-house) APD file format was initially developed to store Affymetrix probe-level data, e.g. normalized CEL intensites. Chip types can be added to APD file and similar to methods in the affxparser package, this package provides methods to read APDs organized by units (probesets). In addition, the probe elements can be arranged optimally such that the elements are guaranteed to be read in order when, for instance, data is read unit by unit. This speed up the read substantially.	100% R	beta
aroma.Base	A generic R plugin dispatcher for BASE. http://www.maths.lth.se/help/R/aroma.Base/	100% R	beta
aroma.light	Light-weight methods for normalization and visualization of microarray data using only basic R data types. Methods for microarray analysis that take basic data types such as matrices and lists of vectors. These methods can be used standalone, be utilized in other packages, or be wrapped up in higher-level classes. The purpose of this package is support higher-level methods in aroma, but also to make its method available to the BASE and Bioconductor community.	Bioconductor; 100% R	beta & stable
R.audio	Audio and sound classes.	100% R DEPRECATED	trial: alpha & beta
R.basic	Stand-alone basic functions. Many of these functions have now been moved to R.utils and aroma.light.	100% R, contains deprecated methods	beta & stable
R.batch	Methods to run batch jobs in R. The package provides methods to batch run R scripts in a straightforward manner. A job batch is defined by a root directory containing subdirectories todo/, running/, finished/, failed/, src/ among others. A job to be processed is a directory placed in todo/. A job is processed by moving it to running/, then sourcing the R files in src/ (containing common code) and then the ones in the job directory (contain job specific code), in which a function onRun() has to be defined, which is then called. Succesfully jobs end up in finished/ whereas failed jobs in failed/.	100% R	beta
R.cache	Fast and light-weight caching of objects. Methods for caching arbitrary R objects on persistent media. Objects can be loaded and saved stratified on a set of hashing objects.	CRAN, 100% R	beta
R.colors	Color and color filter classes.	100% R	trial: alpha & beta
R.graphics	Graphics classes.	100% R, contains deprecated methods	alpha & beta
R.huge	Methods for accessing huge amounts of data. The package currently provides a class representing an array where the data is stored in a binary format on the local file system. This way the size limit of the data is set by the file system and not the memory.	CRAN, 100% R	trial: alpha & beta
R.image	Classes and methods for bitmap images.	100% R	trial: alpha & beta
R.io	File, connection and I/O classes. Many of the methods have been moved to R.utils and R.rsp.	100% R, contains deprecated methods	alpha & beta
R.matlab	Read and write of MAT files together with R-to-Matlab connectivity. http://www.maths.lth.se/help/R/R.matlab/	CRAN, 100% R	beta & stable
R.oo	R object-oriented programming with or without references. In principal all package listed here depends on this package. http://www.maths.lth.se/help/R/R.oo/	CRAN, 100% R	stable
R.rsp	R Server Pages and Light-weight HTTP daemon (server). http://www.maths.lth.se/help/R/R.rsp/	CRAN; 100% R	beta & stable
R.utils	Various programming utilities. Package provides utility classes and methods useful when programming in R and developing R packages.	CRAN, 100% R	alpha, beta & stable

The Application Programming Interface (API) of the classes and methods in alpha (pre-beta) and beta stages may change without notice. Some packages are just trial package, which means that I do not use them much in my own work.

Packages are moved to CRAN or Bioconductor when I consider them to be stable enough.

In addition to all of the above packages, I have also released a few other packages over the years. If not stated elsewise, these should be considered deprecated. It is typically the case that the methods in such packages have been moved or reimplemented in one of the above packages.

Requirements, Dependencies and Package versions

All of the above packages required that you have a very recent version of R installed. If you have problem with some my packages and you run a more than 10 months old version of R, please do not expect me to troubleshoot it for you. To find out what the latest version is for a given package or to see what packages a certain package depends on, please see the PACKAGES file.

Installation and updates

To install the latest (=devel) version of all the above package, do

source("http://www.braju.com/R/hbLite.R")
hbLite()

To install some of the packages including all packages they depend on, do

hbLite(c("R.oo", "R.utils"))

Aditional options: To install the CRAN versions, if they exists, add argument CRAN=TRUE. To force a re-installation of some of the packages, add arguments force=TRUE.

If the above does not work for you, please see Frequently Asked Questions below.

Source Code

To get hold off the source code of the latest version of a package (or a bundle), do

source("http://www.braju.com/R/hbLite.R")
hbGet("aroma.Base")

Contributions

Contributions are appreciated. Please make sure you are working with the latest version. Minor contributions will be acknowledged and major will make you a co-author.

Frequently Asked Questions

FAQ #1. hbLite() fails to install your packages

Make sure you call the above source() command too. If for some reason it still wont work, see FAQ #2 below. If that won't work, please check you firewall and make sure your proxy settings are correct. If you are on Windows, please see Section "2.19 The Internet download functions fail" in "R for Windows FAQ", which you get by calling help.start().

FAQ #2. Can I install your package using install.packages()?

Sure. The repository is at http://www.braju.com/R/repos/, e.g.

install.packages(c("R.oo", "R.utils"), contriburl="http://www.braju.com/R/repos/")

Make sure to all install required package first (hbLite() takes care of this for you).

FAQ #3. Can I install your package using Rgui?

No, except for the few of them that are on CRAN.

FAQ #4. How do I update all of your packages?

To update all installed packages (from here) and their required packages, do:

hbLite()

The classic way to do this is to call update.packages(contriburl="http://www.braju.com/R/repos/").

FAQ #5. Where is your bundle R.classes?

The R.classes bundle is deprecated since 2005-07-21 (v0.62) in favor of standalone packages. The bundle was created so that the aroma package could be installed in a simple way. It contains packages R.oo, R.audio, R.basic, R.batch, R.colors, R.graphics, R.image, R.io, R.lang, R.matlab, R.ui as they were at the time. However, now since some of these packages are now used not only by aroma users, I have decided to discontinue the bundle.

If you still insist to install the bundle, do

install.packages("R.classes", contriburl="http://www.braju.com/R/repos/R.classes")

FAQ #6. How do I get hold of previous versions of a package?

To get hold off the source code or a binary installation of an older version of a package (or a bundle), do

source("http://www.braju.com/R/hbLite.R")
hbGet("aroma.Base", "0.4.1", "tar.gz")   # Source
hbGet("aroma.Base", "0.4.1", "zip")      # Windows binary

If the above for some reason does not work, you can download the files manually by for example http://www.braju.com/R/repos/aroma.Base_0.4.1.tar.gz.

The web server does unfortunately not allow us to list existing packages, so for now you have to set the version number by trial and error to see if it exists.

Contact information

To contact me regarding one of the package, please email me at henrikb @ braju . com.