Home

Build Info

To build this vignette, run

workflowr::wflow_publish("analysis/index.Rmd") or rmarkdown::render("analysis/index.Rmd")

from your R console.

library(workflowr);

Package Info

For package version and information, see Session Information at the end of this documentation.

library(mr.ash.alpha); # package implementing proposed methods in the paper
library(Matrix); library(mvtnorm) # miscellaneous packages for matrices
library(glmnet); library(ncvreg); library(L0Learn); # penalized regression methods
library(varbvs); library(susieR); library(BGLR); # Bayesian methods
library(ggplot2); library(cowplot); # plotting

Tutorial for the Repository

First, load the main functions.

source('code/method_wrapper.R')
source('code/sim_wrapper.R')

One can generate the data as follows.

data = simulate_data(n = 100, p = 200, s = 5,
                     design = "indepgauss", # "equicorrgauss",
                     signal = "normal", # "lap", "t", etc.
                     pve = 0.5)
str(data)

List of 6
 $ X     : num [1:100, 1:200] 1.291 0.224 1.232 -0.869 -1.274 ...
 $ X.test: num [1:100, 1:200] -0.0288 -0.1941 -0.9014 1.3133 -0.8224 ...
 $ y     : num [1:100, 1] -2.42 -6.62 -4.42 2.53 7.11 ...
 $ y.test: num [1:100, 1] -0.277 -4.877 7.563 3.549 3.187 ...
 $ sigma : num 2.62
 $ beta  : num [1:200] 0 0 0 0 0 0 0 0 0 0 ...

One can run each method by calling the wrapper. For instacne, for fitting mr.ash simply call fit.mr.ash.

method_list  = c("mr.ash","varbvs","bayesb","blasso","susie","enet","lasso","ridge","scad2","mcp2","l0learn")
fit.function = get(paste0("fit.", method_list[1], sep = ""))

fit          = fit.function(data$X, data$y, data$X.test, data$y.test)

Each fit.function runs under the default setting of each method.

It can accept non-default argument, for instance, fit.susie(..., L = 20).

Reproduce the results for the paper

To reproduce the results in the paper, run the following code.

You need to install workflowr R package, or replace workflowr::wflow_publish by rmarkdown::render.

In analysis folder of the repository, *_Sourcecode.Rmd files (e.g. Figure2_Sourcecode1.Rmd) contain source codes.

Each *_Sourcecode.Rmd file requires minutes/hours/days to be rendered by MacBook Pro (15-inch, 2016) 2.6 GHz Intel Core i7 16GB RAM.

filelist          = paste("analysis/", list.files("analysis", pattern = "*_Sourcecode.Rmd"), sep = "")
for (i in 1:length(filelist)) {
  workflowr::wflow_publish(filelist[i])
  cat("file",i,": ",'"',filelist[i],'"'," is published\n",sep = "")
}

Create Figures in the paper

To generate figures in the paper, run the following code.

You need to have results from the code in the above chunk.

filelist          = paste("analysis/", list.files("analysis", pattern = "*_Plot.Rmd"), sep = "")
for (i in 1:length(filelist)) {
  workflowr::wflow_publish(filelist[i])
  cat("file",i,": ",'"',filelist[i],'"'," is published\n",sep = "")
}

Session Information

The below is for providing R version and package information.

sessionInfo()

R version 3.5.3 (2019-03-11)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS High Sierra 10.13.3

Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] cowplot_0.9.4      ggplot2_3.2.1      BGLR_1.0.8        
 [4] susieR_0.7.1       varbvs_2.6-5       L0Learn_1.1.0     
 [7] ncvreg_3.11-1      glmnet_2.0-16      foreach_1.4.8     
[10] mvtnorm_1.0-12     Matrix_1.2-17      mr.ash.alpha_0.1-4
[13] workflowr_1.3.0   

loaded via a namespace (and not attached):
 [1] tidyselect_0.2.5    xfun_0.6            purrr_0.3.3        
 [4] reshape2_1.4.3      lattice_0.20-38     colorspace_1.4-1   
 [7] htmltools_0.4.0     yaml_2.2.0          rlang_0.4.2        
[10] pillar_1.4.3        glue_1.3.1          withr_2.1.2        
[13] RColorBrewer_1.1-2  lifecycle_0.1.0     plyr_1.8.5         
[16] stringr_1.4.0       munsell_0.5.0       gtable_0.3.0       
[19] codetools_0.2-16    evaluate_0.13       latticeExtra_0.6-28
[22] knitr_1.22          Rcpp_1.0.3          scales_1.1.0       
[25] backports_1.1.5     truncnorm_1.0-8     fs_1.3.0           
[28] digest_0.6.23       stringi_1.4.5       dplyr_0.8.3        
[31] nor1mix_1.2-3       grid_3.5.3          rprojroot_1.3-2    
[34] tools_3.5.3         magrittr_1.5        lazyeval_0.2.2     
[37] tibble_2.1.3        crayon_1.3.4        whisker_0.3-2      
[40] pkgconfig_2.0.3     assertthat_0.2.1    rmarkdown_1.12     
[43] iterators_1.0.12    R6_2.4.1            git2r_0.25.2       
[46] compiler_3.5.3     

Session information

sessionInfo()

R version 3.5.3 (2019-03-11)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS High Sierra 10.13.3

Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] cowplot_0.9.4      ggplot2_3.2.1      BGLR_1.0.8        
 [4] susieR_0.7.1       varbvs_2.6-5       L0Learn_1.1.0     
 [7] ncvreg_3.11-1      glmnet_2.0-16      foreach_1.4.8     
[10] mvtnorm_1.0-12     Matrix_1.2-17      mr.ash.alpha_0.1-4
[13] workflowr_1.3.0   

loaded via a namespace (and not attached):
 [1] tidyselect_0.2.5    xfun_0.6            purrr_0.3.3        
 [4] reshape2_1.4.3      lattice_0.20-38     colorspace_1.4-1   
 [7] htmltools_0.4.0     yaml_2.2.0          rlang_0.4.2        
[10] pillar_1.4.3        glue_1.3.1          withr_2.1.2        
[13] RColorBrewer_1.1-2  lifecycle_0.1.0     plyr_1.8.5         
[16] stringr_1.4.0       munsell_0.5.0       gtable_0.3.0       
[19] codetools_0.2-16    evaluate_0.13       latticeExtra_0.6-28
[22] knitr_1.22          Rcpp_1.0.3          scales_1.1.0       
[25] backports_1.1.5     truncnorm_1.0-8     fs_1.3.0           
[28] digest_0.6.23       stringi_1.4.5       dplyr_0.8.3        
[31] nor1mix_1.2-3       grid_3.5.3          rprojroot_1.3-2    
[34] tools_3.5.3         magrittr_1.5        lazyeval_0.2.2     
[37] tibble_2.1.3        crayon_1.3.4        whisker_0.3-2      
[40] pkgconfig_2.0.3     assertthat_0.2.1    rmarkdown_1.12     
[43] iterators_1.0.12    R6_2.4.1            git2r_0.25.2       
[46] compiler_3.5.3