R package for fine mapping genetic associations in imputed GWAS data, detailed in
Wallace et al. (2015) Dissection of a Complex Disease Susceptibility Region Using a Bayesian Stochastic Search Approach to Fine Mapping. PLoS Genet 11(6): e1005272. doi: 10.1371/journal.pgen.1005272
which is available (open access) at http://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1005272
install.packages("devtools") # if you don't already have the package
library(devtools)
install_github("chr1swallace/GUESSFM")GUESSFM is purely an R package and so is platform independent. It depends on the software GUESS, which is available at http://www.bgx.org.uk/software/guess.html, but can be more easily installed via the R package R2GUESS, which is a dependency of GUESSFM. GUESS is described in the paper http://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1003657
To install, you may first need some R package dependencies, the packages VGAM, reshape, ggplot2, grid, ggbio, snpStats, parallel, R2GUESS. E.g., if you don’t have ggpplot, from inside R, do
install.packages("ggplot") and repeat for each package you don’t have. NB, R2GUESS contains some library dependencies that required me to do (on Ubuntu):
sudo apt-get install libgsl0-dev
sudo apt-get install libglu1-mesa-devSome packages (e.g. ggbio, snpStats) are from Bioconductor. For these, you need to do
## try http:// if https:// URLs are not supported
source("https://bioconductor.org/biocLite.R")
biocLite("ggbio")
biocLite("snpStats")Then, to install GUESSFM direct from github, do
install.packages("devtools") # if you don't already have the package
library(devtools)
install_github("chr1swallace/GUESSFM")If you want the vignettes to build as well, this should work, but doesn’t (even though they build locally):
install_github("chr1swallace/GUESSFM", build_vignettes=TRUE)Instead, see the vignettes direct from github:
To load the GUESSFM library and see the vignette, do
library(GUESSFM)
vignette(package="GUESSFM")If you run into any problems, please do raise an issue on the github by clicking “Issues” on the top right of this page, and then the green “New Issue” button. Raising issues is better than email, because I will have a reminder of what I need to fix every time I see the page, and your issue may help others. If you still prefer, you could email me at cew54 at cam.ac.uk.