Name Mode Size
R 040000
demo 040000
inst 040000
man 040000
src 040000
tests 040000
vignettes 040000
DESCRIPTION 100644 2 kb
LICENSE 100644 0 kb
NAMESPACE 100644 1 kb
README 100644 3 kb
README
********************************************* * * * PAA (PROTEIN ARRAY ANALYZER) README * * * ********************************************* --------------------------------- | 1. PAA - PACKAGE INSTALLATION | --------------------------------- The recommended way to install PAA is to type the commands described below in the R console: > # only if you install a Bioconductor package for the first time > source("http://www.bioconductor.org/biocLite.R") > # else > library("BiocInstaller") > biocLite("PAA", dependencies=TRUE) This will install PAA including all dependencies. --------------------------------- | 2. RANDOM JUNGLE INSTALLATION | --------------------------------- PAA has an external dependency that is needed to provide full functionality. This external dependency is the free C++ software package Random Jungle (RJ) that can be downloaded from http://www.randomjungle.de/. PAA will be usable without RJ. However, it needs this package for random jungle recursive feature elimination (RJ-RFE) provided by the function selectFeatures(). Please follow the following instructions for your OS to install RJ properly on your machine. 2.1 Windows ----------- (a) Go to "http://www.randomjungle.de/". (b) Download and unzip the appropriate Windows archive (32 bit 0r 64 bit) containing precompiled RJ binaries. (c) Make RJ available for PAA by adding the path to the binaries (e.g., "C:\rjungle\programs\rjungle\bin") to the PATH variable. (d) As a quick test for a successful installation type "rjungle --version" or "rjungle --help" in the Windows Command Prompt. 2.2 Linux --------- (a) Go to "http://www.randomjungle.de/". (b) Download and extract the appropriate Linux tarball (32 bit 0r 64 bit) containing precompiled RJ binaries. (c) Copy the binaries "rjungle" and "rjungle_static" to "/usr/local/bin" to make RJ available for PAA. (d) As a quick test for a successful installation type "rjungle --version" or "rjungle --help" in the Linux console. 2.3 Mac ------- Unfortunately in the medium term there will be no Mac binaries available on "http://www.randomjungle.de/". But it is planned in the long term (according to the RJ developers). As a consequence RJ has to be compiled from the source code. (a) Install gcc (download from "https://gcc.gnu.org/") via "configure", "make" and "make install". (b) Download boost(e.g., "boost_1_55_0.tar.bz2" from "http://www.boost.org/") and install it via "tar --bzip2 -xf /path/to/boost_1_55_0.tar.bz2". (c) Install gsl (download from "http://www.gnu.org/software/gsl/") via "configure", "make" and "make install". (d) Download "randomjungle-master.zip" from "https://github.com/insilico/randomjungle" via the "Download ZIP" button. (e) Unzip "randomjungle-master.zip" then cd into the new randomjungle-master order and "make clean". (f) "./configure CXX='/usr/local/bin/gcc' CXXFLAGS='-fpermissive -lstdc++' --with-boost='/usr/local/boost_1_55_0'" (-lstdc++ is needed to solve possible linking errors). (g) "make". (h) "make install". (i) By default the binaries "rjungle" and "rjungle_static" will be installed in "/usr/local/bin" and should be available for PAA immediately. (j) As a quick test for a successful installation type "rjungle --version" or "rjungle --help" in the Mac terminal.