Example for Bayesian variable selection book chapter:

Variable selection for hierarchically-related outcomes: models and algorithms

Data

The genotyping data are protected. We are therefore using a synthetic dataset emulating the real data. The expression and replicated genotyping data are in data/replicated_data.RData. The ready-to-use data with simulated genetic associations are in data/prepared_data.RData.

Important note: these are large files which are stored using Git Large File Storage. To clone these files along with the repository, please install Git LFS, e.g., using Homebrew:

brew install git-lfs

and, within the cloned repository, initialise it for your account by using:

git lfs install

and retrieve all large files:

git lfs pull

(Alternatively, since there are only two such files, they can be downloaded manually from the Github interface.)

The file data/prepared_data.RData is obtained by updating the real expression data to simulate genetic associations between the synthetic genotyping data and the transcript levels. This last step is obtained by running the R file scripts/prepare_data.R after installing the R package echoseq:

if(!require(remotes)) install.packages("remotes")
remotes::install_github("hruffieux/echoseq")

Algorithm

The package atlasqtl used for the analysis may also be installed with:

if(!require(remotes)) install.packages("remotes")
remotes::install_github("hruffieux/atlasqtl")

eQTL analysis

The eQTL analysis can be run using the Rmarkdown script: scripts/atlasqtl_example.Rmd. This file also provides step-by-step guidance for the use and settings of the atlasqtl for our example.

Issues

To report an issue, please use the issue tracker at github.com.