This repository contains a software mvBIMBAM, a version of BIMBAM for multivariate association analysis, implementing a Bayesian approach for genetic association analysis of multiple related phenotypes.
We adapted source codes in BIMBAM to deal with multiple related phenotypes.
mvBIMBAM is a free software, you can redistribute it and/or modify it under the terms of the GNU General Public License (GPLv3+).
The GNU General Public License does not permit this software to be redistributed in proprietary programs.
This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Binary executable file for Linux is in the mvBIMBAM/bin/ directory (complied on 06/17/2014).
A zip file of binary executable file, source code, and example input files for MAC is in mvBIMBAM\forMAC directory (tested on OS 10.6, but should also work under other 10.x)
cd into the mvBIMBAM/ directory
make all
Then, binary executable file will be created in the mvBIMBAM/ directory.
User manual is in the mvBIMBAM/doc/ directory.
Report bugs as issues on this repository or email the mailing list
- H Shim et al (2015). A multivariate genome-wide association analysis of 10 LDL subfractions, and their response to statin treatment, in 1868 Caucasians. PLoS ONE 10(4): e0120758.
- M Stephens (2013). A Unified framework for association analysis with multiple related phenotypes. PLoS ONE 8(7): e65245.
- The CAP genotypes are in dbGAP : link
- (1) Normalized averages and differences of LDL subfractions, (2) genotype data for selected SNPs, and (3) summary statistics (effect size estimate and standard error) for each pair of a single SNP and a single phenotype are available here.
Heejung Shim (University of Chicago)