[Genabel-commits] r1155 - pkg/OmicABEL
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Sun Mar 17 16:56:41 CET 2013
Author: dfabregat
Date: 2013-03-17 16:56:41 +0100 (Sun, 17 Mar 2013)
New Revision: 1155
Added:
pkg/OmicABEL/DISCLAIMER.Linux
pkg/OmicABEL/HowTo
pkg/OmicABEL/INSTALL
pkg/OmicABEL/README
Log:
Adding documentation files
Added: pkg/OmicABEL/DISCLAIMER.Linux
===================================================================
--- pkg/OmicABEL/DISCLAIMER.Linux (rev 0)
+++ pkg/OmicABEL/DISCLAIMER.Linux 2013-03-17 15:56:41 UTC (rev 1155)
@@ -0,0 +1,6 @@
+The Linux precompiled binary for CLAK-GWAS provided in this package
+includes libraries distributed under their respective licenses.
+This is the case for the Intel (R) Math Kernel Library (MKL).
+Under the terms of the Intel MKL License we redistribute
+computational portions of Intel MKL with our software to
+improve the user experience.
Added: pkg/OmicABEL/HowTo
===================================================================
--- pkg/OmicABEL/HowTo (rev 0)
+++ pkg/OmicABEL/HowTo 2013-03-17 15:56:41 UTC (rev 1155)
@@ -0,0 +1,79 @@
+1. Who is Who
+--------------
+
+In the bin/ directory you can find the following executables:
+
+* CLAK-GWAS: the program to run GWAS analyses (through CLAK-Chol or CLAK-Eig)
+* reshuffle: the program to extract the output of CLAK-GWAS into text format
+
+The output produced by CLAK-GWAS is kept in a compact binary format
+for performance reasons. The user can then use "reshuffle" to
+extract the desired information. Since the output may be very large,
+reshuffle allows the selection of specific ranges of SNPs and traits.
+
+
+2. Preparing your data
+-----------------------
+
+Currently, CLAK-GWAS only accepts as input data files in DenABEL
+(filevector) format.
+In the tutorial available in the project's webpage:
+
+http://www.genabel.org/packages/OmicABEL
+
+you can find an example of how to transform your data into filevector.
+If you need help, don't hesitate contacting us or ask for help at the
+GenABEL project forum
+
+http://forum.genabel.org
+
+
+3. Running GWAS analyses
+-------------------------
+
+The example in the tutorial also provides a basic example on using OmicABEL
+to run your GWAS analyses. Here we detail the options of CLAK-GWAS.
+The complete list of options for CLAK-GWAS is avaliable through the command
+
+./CLAK-GWAS -h
+
+Usage: CLAK-GWAS [options]
+
+Following options may be given:
+
+ -var [chol | eigen] Default is chol.
+ -cov base path to covariates file
+ -phi base path to kinship matrix file
+ -snp base path to SNPs file
+ -pheno base path to phenotypes file
+ -out base path to output file
+ -nths # Default is 1 thread.
+ -thres # Default is 95%.
+ -h Show this help and exit
+
+* The option "-var" selects the algorithm to use:
+ "chol" forces CLAK-GWAS to use the CLAK-Chol algorithm (recommended for
+ single-phenotype GWAS).
+ "eigen" forces CLAK-GWAS to use the CLAK-Eig algorithms (recommended for
+ multiple-phenotype GWAS).
+
+* "-cov", "-phi", "-snp", "-pheno" are used to indicate the input files
+ for the analysis. If the covariates are stored in the pair of
+ file "covs.fvd" and "covs.fvi", the option -cov
+ is used as: "-cov covs". Same for the other three options.
+
+* "-out" is used to indicate the output file(s)
+
+* "-nths" allows the user to indicate how many cores to use.
+
+* "-thres" is the acceptable proportion of missing values
+ (these are imputed by means, therefore our suggestion
+ is to keep this threshold at 95 at minimum).
+
+
+Options without a default value (-cov, -phi, -snp, -pheno, -out)
+are mandatory.
+
+
+For a detail description of the options for "reshuffle", please
+see src/reshuffle/README.txt
Added: pkg/OmicABEL/INSTALL
===================================================================
--- pkg/OmicABEL/INSTALL (rev 0)
+++ pkg/OmicABEL/INSTALL 2013-03-17 15:56:41 UTC (rev 1155)
@@ -0,0 +1,29 @@
+1. Pre-requisites
+------------------
+
+ Our software requires the availability of the BLAS and LAPACK
+ libraries. If you do not have them installed in your system,
+ you can freely downloaded them from:
+
+ * BLAS: http://xianyi.github.com/OpenBLAS/
+ * LAPACK: http://netlib.org/lapack/#_software
+
+ Their installation is typically rather straightforward.
+
+
+2. Compiling
+-------------
+
+ The file make.inc sets a number of environment variables
+ necessary for the compilation. We provide a number of examples
+ of this file in the inc/ directory, for different combinations
+ of architectures, compilers, and libraries. You can choose the
+ most suitable and overwrite the file make.inc with it.
+ For instance, if you chose to install the BLAS and LAPACK libraries
+ linked above in a linux machine, select the file "make.inc.gnu-goto".
+ You will need to set the paths to the aforementioned libraries
+ in your system.
+
+ Once the make.inc file is ready, you can compile the code.
+ Simply type "make" and it will create the binaries "CLAK-GWAS"
+ and "reshuffle" in the bin/ directory.
Added: pkg/OmicABEL/README
===================================================================
--- pkg/OmicABEL/README (rev 0)
+++ pkg/OmicABEL/README 2013-03-17 15:56:41 UTC (rev 1155)
@@ -0,0 +1,52 @@
+CLAK-GWAS is a software for performing Genome-Wide Association Studies
+(GWAS). It provides a high-performance implementation of two
+algorithms, CLAK-Chol and CLAK-Eig, for GWAS involving single and
+multiple phenotypes, respectively.
+
+0. LICENSE
+-----------
+
+OmicABEL is distributed under the GNU GENERAL PUBLIC LICENSE v.3 or later
+A copy of the license is available in the file ''COPYING''
+
+
+1. INSTALLATION
+----------------
+
+Please see the INSTALL file (only necessary if you didn't download a
+precompiled version).
+
+
+2. USING CLAK-GWAS
+-------------------
+
+Please see the HowTo file.
+
+
+3. KNOWN ISSUES
+----------------
+
+3.1 For the sake of ease-of-use, CLAK-GWAS automatically performs
+ a number of system checks and tries to maximize performance.
+ Should you experience any unexpected behavior, please contact us
+ ( fabregat [at] aices [dot] rwth-aachen [dot] de ); we will attend
+ your request as soon as possible.
+
+3.2 Currently, we only support input data in DenABEL (filevector) data format
+ and double precision datatype. We are working on the support for
+ more datatypes.
+
+
+4. CONTACT
+-----------
+
+For any question, request, suggestion, or comment, please free to contact us:
+
+Diego Fabregat Traver, fabregat [at] aices [dot] rwth-aachen [dot] de
+Paolo Bientinesi, pauldj [at] aices [dot] rwth-aachen [dot] de
+Yurii Aulchenko, yurii [dot] aulchenko [at] gmail [dot] com
+
+You can also seek help in our forum: http://forum.genabel.org/
+
+
+Happy GWASing!
More information about the Genabel-commits
mailing list