Installation

Note

Want to try ecoevolity without installing it? Try out our Docker image. If you run into trouble below building and installing ecoevolity, the Docker image provides an alternative.

Prerequisites

Compiling ecoevolity requires CMake and a new-ish C++ compiler (one that supports the C++11 standard).

We also strongly recommend using Git to acquire the source code. If you use Git, the ecoevolity tools will report the version of ecoevolity you are using much more precisely, which will make your work more reproducible.

While not required, the pycoevolity Python package can be useful for summarizing the output of ecoevolity. Pycoevolity will be used in tutorials. See below for how to install pycoevolity.

Linux

If you are on a Linux machine, use your package manager to update/install g++, CMake, and Git. On Ubuntu (or other Debian-based Linux distribution) you can simply use:

sudo apt-get install cmake g++ git

Mac

If you are on a Mac, you can install Xcode command line tools (that’ll get you a C++ compiler and Git) and download and install CMake from https://cmake.org/download/. Download the .dmg file for the latest version for Mac. Once installed, open CMake, and from the “Tools” menu click “How to Install For Command Line Use.” This should pop up a window that has a line like:

sudo "/Applications/CMake.app/Contents/bin/cmake-gui" --install

The line might be slightly different on your Mac. Copy that line from the CMake pop up window (not the line above), and paste it into the command line (Terminal).

Windows

We have not compiled ecoevolity in Windows. It’s possible to do, but I suspect it will take a fair bit of tweaking of the CMake configuration. If you have a Windows machine, and you are not experienced with compiling C/C++ code in Windows, you have some options:

  1. If you have Windows 10, you can install Ubuntu. Once installed, then you can follow the instructions above for Linux.

  2. Try running our Docker image (see below).

Getting the ecoevolity source code

If you have Git installed, download the ecoevolity repository with the clone command:

git clone https://github.com/phyletica/ecoevolity.git

If you prefer not to use Git, you can download an archive of the source code here. Once downloaded, you’ll need to unzip the archive.

Basic build and install

Next, move into the downloaded directory and run the build script:

cd ecoevolity
./build.sh

If the build was successful, the ecoevolity executables should now be in the ./build/bin directory, and you should be able to run:

./build/bin/ecoevolity -h

and see the ecoevolity help menu, the beginning of which should looks something like:

======================================================================
                              Ecoevolity
                  Estimating evolutionary coevality
        Version 0.2.0 (dev d4a4d48: 2018-05-23T15:22:40-05:00)
======================================================================

Usage: ecoevolity [OPTIONS] YAML-CONFIG-FILE

Ecoevolity: Estimating evolutionary coevality

Options:
  --version             show program's version number and exit
  -h, --help            show this help message and exit

The executables in ./build/bin are ready to use, but you’ll probably want to put them in your PATH (a list of directories that your shell looks in to find the commands you type on the command line). You can do this via:

sudo cp ./build/bin/* /usr/local/bin

If this worked, you’re good to go; you can try ecoevolity -h to be sure.

If it didn’t work, you probably don’t have admin privileges. If so, you can create a bin folder in your home folder and put the tools there:

mkdir -p "${HOME}/bin"
cp ./build/bin/* "${HOME}/bin"

Then, you can add this directory to your PATH (if it’s not already there; you can check with echo $PATH):

export PATH="${PATH}:${HOME}/bin"

Note, this update to PATH is only for your current terminal window. If you want this to be permanent (work for all future terminal windows), add export PATH="${PATH}:${HOME}/bin" to your .bashrc or .bash_profile file in your home directory.

Install during build

If you want to build and install in one go, you just need to specify where you want the installation to go, for example:

sudo ./build.sh --prefix /usr/local

Building the threaded version

If you want to install a version of ecoevolity that performs the likelihood calculations across multiple threads, you just need to add the --threads flag:

./build.sh --threads

In my opinion, you’re usually better off running multiple independent chains rather than multithreading, but the option is there.

Installing pycoevolity

Pycoevolity is a Python package for summarizing the output of ecoevolity. It should work with Python 2 or 3. If you have Python and pip installed, you can install Pycoevolity via:

pip install git+git://github.com/phyletica/pycoevolity.git

If this isn’t working, try the manual installation instructions here. Also, pycoevolity uses the R packages ggplot2 and ggridges for creating some plots. So, if you want plotting by pycoevolity to be fully functional, and you don’t already have R installed, you’ll need to install it. Once R is in place, you can install the packages from the R prompt using::

install.packages(c("ggplot2", "ggridges"))

Using ecoevolity without installing

Docker provides a nice way of sharing lightweight containers that act like a virtual machine. We have created a Docker container with ecoevolity built in. To get started, you first need to install Docker. To do so, go to https://docs.docker.com/install and scroll down and click on your platform under “Supported platforms.” If you’re on a Mac or Windows machine, might need to sign up for a free Docker account to download Docker Desktop. Once Docker is installed and running pull down our Docker image:

docker pull phyletica/ecoevolity-docker

Note

Depending on your system and how Docker is configured, you may need to use sudo to run Docker commands. If you received a “permission denied” message when you ran the command above, try:

sudo docker pull phyletica/ecoevolity-docker

This download could take several minutes depending on your internet connection. After it completes, run and enter the docker container:

docker run -it phyletica/ecoevolity-docker bash

Note

Again, you might need to prefix this command with sudo.

That’s it, you are now in a virtual container with a fully functioning ecoevolity ecosystem (ecoevolity and pycoevolity are installed, along with example data). Try typing:

ecoevolity -h

This should display the ecoevolity help menu. Next, cd into the example data directory:

cd ecoevolity-example-data
ls

There you will find an ecoevolity configuration file and nexus-formatted data files. Go ahead and run an ecoevolity analysis:

ecoevolity --relax-missing-sites --relax-triallelic-sites --ignore-data ecoevolity-config.yml

To exit the container, simply type:

exit

Docker will keep the ecoevolity image around, so you can always jump back in anytime via:

docker run -it phyletica/ecoevolity-docker bash

However, any files you created on your last visit will be gone. So, if you want to analyze your data and keep the results around, cd to the directory where you want to run ecoevolity, then jump into the Docker container using:

docker run -v "$(pwd)":/portal -it phyletica/ecoevolity-docker bash

Then, once inside, type:

cd portal
ls

You should see the files that were in the directory on your computer. Now you can run ecoevolity on data in this directory, and all output files will be on your computer when you exit the container (magic!).

Getting some example data

You can get an example of an ecoevolity configuration file and nexus-formatted data files from one of our GitHub repos:

git clone https://github.com/phyletica/ecoevolity-example-data.git

If you prefer not to use Git, you can download an archive of the example data here.

Note

If you are using the Docker image, the example data are included in the container. But, for the tutorials, it will be helpful to follow the instructions above to get a copy on your computer, outside of the container.