Install EnzyHTP
This guide leads you through steps of installing EnzyHTP
I. Make a new environment
conda create --name enzy_htp python=3.9
conda activate enzy_htp
enzy_htp and we will install python package EnzyHTP needs in it.conda activate enzy_htpII. Install EnzyHTP
There are 2 ways of installing EnzyHTP: 1) from conda 2) from source
Note
Make sure you are under the correct environment when doing this step
a. Install from conda
Currently we recommend installing from conda.
conda install enzy_htp -c yang_lab -c conda-forge
b. Install from source
Obtain the source code
EnzyHTP is shared-source on GitHub.
git clone https://github.com/ChemBioHTP/EnzyHTP.gitThis will generate a folder name “EnzyHTP” containing the source code under where you run this command.
Note
Note that the source code version when writing this tutorial is stored as a release too. (try this version if this tutorial fails with the most current version)
Install Python packages
cd EnzyHTP # this is what you created via clone in the previous step source dev-tools/conda-install
III. Install External Dependencies
1. Install Amber (optional)
EnzyHTP use Amber as one of the MD engine (currently the only one). You need to install Amber if you want to enable MD/mutation functions in EnzyHTP. AmberTools is free (CPU only) and Amber is charged (GPU support). Most universities have Amber purchased and installed in their local clusters. Contact your local cluster if you don’t know how to load/setup the installed Amber.
As an example, here is the command for loading Amber in our local cluster ACCRE.
source /home/shaoq1/bin/amber_env/amber-accre.sh
2. Install Gaussian16 (optional)
EnzyHTP use Gaussian16 as one of the QM engine (currently the only one). You need to install Gaussian16 if you want to enable QM-based functions in EnzyHTP. Gaussian16 is charged. Most universities have Gaussian16 purchased and installed in their local clusters. Contact your local cluster if you don’t know how to load/setup the installed Gaussian16.
As an example, here is the command for loading Gaussian16 in our local cluster ACCRE.
module load Gaussian/16.B.01
3. Install Multiwfn (optional)
Note
Some HPC may not have the most up-to-date library for running Multiwfn For example, on our local cluster ACCRE, these module needs to be load for Multiwfn to run on a computing node.
module load GCC/6.4.0-2.28 OpenMPI/2.1.1
4. Install Rosetta (optional)
Author: QZ Shao <shaoqz@icloud.com>