1. Install Julia.

2. Install R and RStudio (IDE).

3. Install prerequisite R package JuliaCall and devtools in R.

   install.packages("JuliaCall")
   install.packages("devtools")

   If the error massage Installation failed: Command failed (50) is shown, please install the latest version of devtools as devtools::install_github("hadley/devtools").

4. Install R package JWASr in R

   devtools::install_github("zhaotianjing/JWASr")

Please set up every time you start a new session of R. The set up time is about 10 seconds.

• Mac or Linux users

  JWASr::jwasr_setup()

• Windows users

  # please change to your local path of libjulia.dll
  path_libjulia = "C:/Users/ztjsw/AppData/Local/Julia-1.0.1/bin/libjulia.dll"
  JWASr::jwasr_setup_win(path_libjulia)

  If R session aborted, please click "Start New Session" and set up again.

Note that all data can be found in our subfolder named "data".

library("JWASr")

Please make sure you've already set up.

phenotypes = phenotypes #build-in data

ped_path = "D:\\JWASr\\data\\pedigree.txt" #please change to your local path
pedigree = get_pedigree(ped_path, separator = ',', header = TRUE)  

You can import your own data by read.table().

model_equation = "y1 = intercept + x1*x3 + x2 + x3 + ID + dam";
R = 1.0

model = build_model(model_equation,R)

set_covariate(model, "x1")

G1 = 1.0
set_random(model, "x2", G1)

G2 = diag(2)
set_random_ped(model, "ID dam", pedigree, G2)

G3 = 1.0
geno_path = "D:/JWASr/data/genotypes.txt"  #please change to your local path

add_genotypes(model, geno_path, G3, separator=',', header = TRUE)  

outputMCMCsamples(model, "x2")

out = runMCMC(model, phenotypes, methods = "BayesC", estimatePi = TRUE,
                     chain_length = 5000, output_samples_frequency = 100)

If you change any function in subfolder "R", please run devtools::document() in R to update the package. (under path of JWASr).

In the package, we provide a user friendly Graphical User Interface application, which was build by shiny.

After complete installation above for JWAS, please also install shiny by install.packages("shiny") in R.

In R, please run:

JWASr::runShiny()

• script to install required packages if not installed
• improve GUI interface
• add the complete interface to JWAS.jl