wanwanliang/GPtour

Quick, straight to the point tour on how to implement genomic prediction (GP) in R

GPtour: Genomic Prediction Tour in R

This is a quick, straight to the point tour on how to implement genomic prediction in R using the most popular algorithms (Lasso, Ridge Regression, GBLUP, Single Step, Bayes A,B,C, Multilayer Perceptron, Convolutional Neural Network). We use Rstudio and learnr packages, among other tools.

The main code is in GPtour.Rmd

Requirements

You need to install these packages to be able to run GPtour.Rmd

install.packages("BGLR",repos="https://cran.r-project.org/")  
install.packages("glmnet",repos="https://cran.r-project.org/")  
install.packages("AGHmatrix", repos="https://cran.r-project.org/")  
install.packages('learnr',repos="https://cran.r-project.org/")  
install.packages('downloadthis',repos="https://cran.r-project.org/")  
install.packages("remotes",repos="https://cran.r-project.org/")  
remotes::install_github("rstudio/gradethis") 

On top, for running deepl learning you need to install miniconda, follow instructions in https://keras.rstudio.com/

How to use

only once

install.packages('remotes')
remotes::install_github('miguelperezenciso/GPtour')

to run

# type in your console in Rstudio or regular R terminal
library(GPtour)
library(learnr)
learnr::run_tutorial('tour', 'GPtour')

You may have an error in mac os if running from terminal. Type this

export RSTUDIO_PANDOC=/Applications/RStudio.app/Contents/MacOS/pandoc

Enjoy!