This page contains software installation instructions for Gromacs workshops. When relevant, separate instructions are given for Windows, MacOS and Linux systems.
- For the tutorials on umbrella-sampling ("pulling") simulations and applied-weight histogram methods ("AWH"), you will only need a non-MPI version of GROMACS along with Python3 and some packages. We suggest that you install these using the conda package manager.
- We will also be using Xmgrace (a.k.a. Grace) and optionally VMD.
- The replica-exchange molecular dynamics tutorial requires an MPI version of GROMACS to run production simulations, but you are not strictly required to install it since the tutorial will be about learning the concepts. You can however find general installation instructions for MPI-GROMACS below.
- For the tutorial "Computing trajectories efficiently on GPUs" we will use an external cluster and instructions will be provided during the workshop.
- After you have gone through the installation steps below, please download the tutorial Jupyter notebook, run the notebook from terminal with
jupyter-notebook tutorial.ipynb
, and execute all cells in the notebook to make sure you have everything installed correctly. To download the notebook, visit this page and save it to your hard drive by right-clicking the page and saving it as "tutorial.ipynb".
In the "Running via Binder" section below you will additionally find information on running GROMACS (including an MPI version) in the cloud without any local installation.
We strongly recommend to use (and install if necessary) the Windows Subsystem for Linux, WSL. Inside it you will need Python 3 and the conda Python environment manager. A useful guide to doing this is found at https://github.com/kapsakcj/win10-linux-conda-how-to. Once you have WSL and conda installed, make an environment for the GROMACS tutorials with:
conda create --name gromacs-tutorials -c conda-forge -c bioconda gromacs=2020.4 matplotlib nglview notebook numpy requests pandas seaborn
And activate it by:
conda activate gromacs-tutorials
Install miniconda for Python 3 by following https://docs.conda.io/en/latest/miniconda.html#linux-installers
Once you have conda installed, make an environment for the GROMACS tutorials with:
conda create --name gromacs-tutorials -c conda-forge -c bioconda gromacs=2020.4 matplotlib nglview notebook numpy requests pandas seaborn
And activate it by:
conda activate gromacs-tutorials
Install miniconda for Python 3 by following https://docs.conda.io/en/latest/miniconda.html#macosx-installers
Once you have conda installed, make an environment for the GROMACS tutorials with (note that 2019.1 is the highest available version for MacOS):
conda create --name gromacs-tutorials -c conda-forge -c bioconda gromacs=2019.1 matplotlib nglview notebook numpy requests pandas seaborn
And activate it by:
conda activate gromacs-tutorials
VMD is a powerful molecular visualization package, but it is not required for the GROMACS tutorials since nglview
can be used instead. If you nonetheless want to install VMD, please visit
https://www.ks.uiuc.edu/Development/Download/download.cgi?PackageName=VMD to register, download and install.
Xmgrace is a WYSIWIG 2D plotting tool which will be used in some of the tutorials to plot output data. It can be installed on Linux, MacOS and WSL under Windows.
On Windows-WSL and Linux, type
sudo apt-get install grace
On MacOS, follow these instructions: http://www.phy.ohio.edu/~hadizade/blog_files/XmGrace_Mac.html
The tutorial on replica-exchange MD simulations requires MPI-based GROMACS. Since we will not be running production MD simulations during the tutorial it is not necessary to install an MPI version, and we will be providing reference output files for non-MPI GROMACS as a backup. However, if you want to install MPI-based GROMACS on your own computer we provide here a general installation based on conda which should work on MacOS, Linux and WSL under Windows. If you already have compilers and an MPI library installed on your computer, you can skip the first section below or refer to the official installation instructions (https://manual.gromacs.org/documentation/current/install-guide/index.html).
conda create --name mpi
conda activate mpi
conda install -c conda-forge compilers
conda install -c conda-forge openmpi
conda install -c conda-forge cmake
wget http://ftp.gromacs.org/pub/gromacs/gromacs-2020.4.tar.gz
cd gromacs-2020.4
(or git clone https://gitlab.com/gromacs/gromacs.git; cd gromacs; git checkout v2020.4)
mkdir build
cd build
cmake .. -DCMAKE_C_COMPILER=mpicc -DCMAKE_CXX_COMPILER=mpicxx -DGMX_MPI=ON -DGMX_DOUBLE=OFF -DGMX_BUILD_OWN_FFTW=ON -DREGRESSIONTEST_DOWNLOAD=ON
make
make check
sudo make install
source /usr/local/gromacs/bin/GMXRC
You will now hopefully be able to run gmx_mpi!
As a last resort, if you do not manage to install GROMACS or some of the required dependencies on your local machine, you can run GROMACS (including an MPI version) in the cloud via Binder.
The first step is to launch this repository on Binder by clicking the "launch binder" button at the top of this README file.
This will initiate a cloud instance which installs all the dependencies listed in environment.yml
and start a Jupyter notebook
file browser. You can then create new Jupyter notebooks by clicking "New" on the right side and selecting "Python 3". You will have
access to a gmx
executable inside your notebooks.
You can also compile an MPI version of GROMACS by following these steps in a terminal (click "New" and select "Terminal"):
cd $HOME
wget http://ftp.gromacs.org/pub/gromacs/gromacs-2020.4.tar.gz
tar zxf gromacs-2020.4.tar.gz
cd gromacs-2020.4
mkdir build
cd build
cmake .. -DCMAKE_C_COMPILER=mpicc -DCMAKE_CXX_COMPILER=mpicxx -DGMX_MPI=ON -DGMX_DOUBLE=OFF -DGMX_FFT_LIBRARY=fftpack
make
This procedure will install gmx_mpi
into /home/jovyan/gromacs-2020.4/build/bin
which you can use for MPI-enabled simulations.