tables-figures.Rmd

---
title: "Code to reproduce tables and figures in _Simultaneous SNP selection and adjustment for population structure in high dimensional prediction models_. Bhatnagar et al. (2020). PLOS Genetics."
author: "by Sahir Bhatnagar"
date: "`r Sys.Date()`"
output:
  html_document:
    toc: true
    toc_float: false
    number_sections: true
    toc_depth: 4
header-includes:
   - \newcommand{\bx}{\textbf{\emph{x}}}
   - \newcommand{\by}{\textbf{\emph{y}}}
   - \newcommand{\bX}{\textbf{X}}
   - \newcommand{\bW}{\textbf{W}}
   - \newcommand{\bY}{\textbf{Y}}
   - \newcommand{\bbeta}{\boldsymbol{\beta}}
editor_options: 
  chunk_output_type: console
---

This `.Rmd` file contains the code to reproduce the tables and figures in the paper _Simultaneous SNP selection and adjustment for population structure in high dimensional prediction models_. Bhatnagar et al. (2020). PLOS Genetics. 

# How to compile this document

The easiest way to compile this document is to clone the https://github.com/sahirbhatnagar/ggmix repository or download the zip file https://github.com/sahirbhatnagar/ggmix/archive/master.zip. Open the file [`manuscript/bin/tables-figures.Rmd`](https://github.com/sahirbhatnagar/ggmix/blob/master/manuscript/bin/tables-figures.Rmd) in RStudio and click on the _Knit_ button. If you don't use RStudio, then from the `R` console enter `rmarkdown::render("manuscript/bin/tables-figures.Rmd")`. This assumes that you have installed both the `knitr` and `rmarkdown` packages, and are in the root of the `ggmix` folder. 

The order in which the tables and figures appear follow that of the paper. The `R` scripts for each Table and Figure are in separate `R` files found in the [`manuscript/bin`](https://github.com/sahirbhatnagar/ggmix/tree/master/manuscript/bin) folder. This `.Rmd` document uses [code externalization](https://yihui.org/knitr/demo/externalization/).


# Load Requires Packages


```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE, warning = FALSE, message = FALSE)
options(digits = 4, scipen = 2)
knitr::read_chunk(here::here("manuscript/bin/setup.R"))
knitr::read_chunk(here::here("manuscript/bin/simulation.R"))
knitr::read_chunk(here::here("manuscript/bin/GAW20.R"))
knitr::read_chunk(here::here("manuscript/bin/UKB.R"))
knitr::read_chunk(here::here("manuscript/bin/mice.R"))
```

```{r packages}
```

```{r globals}
```


# Figure 1


```{r plot-kinship-sim}
```


# Figure 2

```{r plot-pc-sim}
```


# Table 1

The simulation study results were run on a cluster using the [simulator](https://cran.r-project.org/package=simulator) package. The simulator files can be found in the [`simulation`](https://github.com/sahirbhatnagar/ggmix/tree/master/simulation) folder:

- [`simulation/main.R`](https://raw.githubusercontent.com/sahirbhatnagar/ggmix/master/simulation/main.R) contains the main file which runs the simulation. This can be thought of as a master file which  in turn calls the methods, model and eval functions  
- [`simulation/model_functions.R`](https://raw.githubusercontent.com/sahirbhatnagar/ggmix/master/simulation/model_functions.R) contains the model functions, i.e., used to simulate the data  
- [`simulation/method_functions.R`](https://raw.githubusercontent.com/sahirbhatnagar/ggmix/master/simulation/method_functions.R) contains the the individual methods which are applied on the simulated data  
- [`simulation/eval_functions.R`](https://raw.githubusercontent.com/sahirbhatnagar/ggmix/master/simulation/eval_functions.R) contains the functions used to evaluate the different methods  



```{r simulation-results}
```


```{r table-of-results}
```


```{r print-sim-table, results='asis'}
```


# Figure 3

```{r UKB-Figure,fig.width=8, fig.height=7}
```


# Table 2

```{r GAW20-prediction-RMSE-activeVariable}
```



# Figure 4

```{r Mice-comparison-fixTPR, fig.width=12, fig.height=10}
```


# S1 Figure

```{r UKB-chromosome-distribution}
```


# S2 Figure

```{r GAW20-R2}
```



# S3 Figure

```{r Mice-R2}
```



# File and Session Information 

  * File creation date: `r Sys.Date()`
  * `r R.version.string`
  * R version (short form): `r getRversion()`
  * Additional session information
  
```{r echo=FALSE}
if (!requireNamespace("devtools")) sessionInfo() else devtools::session_info()
```