The instructions in this blog were written to set up the lab in Rstudio cloud
Setting up your own system
Linux is the operating system of choice to run bioinformatics software. You will need either a computer running linux or or mac os, which has a linux-like environment.
- install anaconda/miniconda
- define imlabtools conda environment how to here, which will install all the python modules needed for this analysis session
- download data and software from Box. This will have copies of all the software repositories and the models
- download metaxcan repo
- download prediction models from predictdb.org
- install R/RStudio/tidyverse package
- install R packages
- git clone https://github.com/hakyimlab/QGT-Columbia-HKI.git
- start Rstudio (if you installed workflowr, you can just open the QGT-Columbia-HKI.Rproj)
#install R packages install.packages("remotes") library(remotes) remotes::install_github("stephenslab/susieR") #gives the most up to date SusieR install.packages("coloc") install.packages("tidyverse") install.packages("R.utils") install.packages("remotes") remotes::install_github("simingz/ctwas", ref = "develop") #cTWAS install install.packages("R.utils") install.packages("remotes") remotes::install_github("simingz/ctwas", ref = "develop") library(ctwas) ## make sure these are installed # library(data.table) # library(BEDMatrix) # library(Rfast) # library(susieR) # library(coloc) #installing base miniconda install.packages("keras") reticulate::install_miniconda("miniconda")
Setting up Conda Environment
#adding path export PATH=$PATH:/cloud/project/miniconda/bin # adding repos/data git clone https://github.com/hakyimlab/QGT-Columbia-HKI-repo.git git clone https://github.com/hakyimlab/MetaXcan.git conda config --append channels conda-forge conda config --append channels bioconda conda env create -f /cloud/project/MetaXcan/software/conda_env.yaml conda activate imlabtools
Set the Ram to 5gb in order to create the environment.
To upload the data. I first downloaded the box folder onto a local machine and then compressed the folders in the box folder to upload onto the server.
Notes: For some reason setting up the environment and uploading the data will continue to use a lot of background RAM set sure to close the project and reopen to close extraneous programs.