Install - nclark-lab/RERconverge GitHub Wiki

We provide instructions below for installation on Ubuntu 14.04 (64 bit), Ubuntu 16.04 (64 bit), CentOS 7 (64 bit), Mac 10.13.1 (64 bit) Win 7 (64 bit), and Win 10 (64 bit). Steps 4 and 5 below are universal, regardless of system. If you meet any problem during the installation, you are welcome to create a new issue and we will help you to resolve it.

Ubuntu 14.04 (64 bit), Ubuntu 16.04 (64 bit) setup

  1. Install R
    • add the following line to /etc/apt/source.list.
      • deb https://cran.rstudio.com/bin/linux/ubuntu trusty/
    • sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E084DAB9
    sudo apt-get update
sudo apt-get install r-base
  2. Install dependcies for devtools
    • sudo apt-get install libssl-dev
    • sudo apt-get install libcurl4-openssl-dev
  3. See installation instructions below

CentOS 7 (64 bit) setup

  1. Install R 
    su
yum install epel-release
yum install R
  2. Install dependcies for devtools 
    su
yum install openssl-devel.x86-64
yum install libcurl-devel.x86_64
  3. See installation instructions below

Mac setup

  1. Install R, Rstudio from the official website
  • https://cran.r-project.org/bin/macosx/
  1. Install Xcode command line tools and appropriate gfortran based on the macOS version
  1. See installation instructions below

Win 7 (64 bit), Win 10 (64 bit) setup

  1. Install R
  2. Install Rtools
  3. See installation instructions below

Universal installation instructions

After you've done the setup for your system, all systems will have the same instructions for the R terminal:

  1. Install R package dependcies 
    devtools
RColorBrewer
gplots
phytools
├──ape
├──maps
├──Rcpp
geiger
knitr
RcppArmadillo
weights
phangorn
  2. Install from Github 
    library(devtools)
install_github("nclark-lab/RERconverge")

That is all! Visit the tutorial vignette to begin using RERConverge.

[Legacy] Mac installation fixes: gfortran, docker, and binaries

These methods are pending deprecation, but we have left them here in case your installation does not go smoothly.

clang replacement

If you have troubles with previously installed clang, please follow this guidance to remove old installation files:

  • https://thecoatlessprofessor.com/programming/cpp/r-compiler-tools-for-rcpp-on-macos/
  • many thanks to @kjf319 for directing us to this great resource.

Gfortran installation

  • If the universal package does not work, you can find alternate gfortran installation files from https://github.com/fxcoudert/gfortran-for-macOS/releases. You need to select the appropriate version based on your macOS and processors.
  • macOS may stop you from installing gfortran. You can type sudo spctl --master-disable in the Terminal and try to install again. After installation, you can restore your setting by typing sudo spctl --master-enable

Use the Docker image

If you have trobules with the installation and you're familiar with Docker, we provide a Docker image (Docker Hub:wem26/rerconverge) for you to use RERconverge. Also we provide a brief instruction on how to set up Docker.

Install from the binary files

You still need to install all R package dependencies separately excluding devtools in order to make the binary work.

  1. Make sure your R version is at least the version under which the binary was compiled. If not, install the latest version of R.
  2. Install the dependent packages by executing the install GitHub function within R as below. The installation may fail to install the entire package, but dependencies will be installed.
library(devtools)
install_github("nclark-lab/RERconverge")
  1. Download the binary from the release page.
  2. Open the Terminal application and change directory to that in which the binary is located. Then, execute the following command in the Terminal: R CMD INSTALL Mac_Big_Sur_R_4.0.0.RERconverge_0.1.0.tgz
  • If you find any installation/source issue, you can check Install FAQ for help. If you don't find a solution, submit an issue ticket.