Installing TinkerModeller

Installation for User

Firstly, you can download it through git or zip file:

git clone git@github.com:WanluLigroupUCSD/TinkerModellor.git

Then, we recommand to use conda to configure the environment:

cd TinkerModeller
conda env create -n tkm -f env.yml
conda activate tkm

A simple way to build TinkerModeller is to use the shell script install.sh

A successful installation will look like the following:

# Installing collected packages: TinkerModeller
# Successfully installed TinkerModeller-1.1

Installation for Developer

After configuring the conda environment

cd TinkerModeller
export TKMROOT=$(pwd)

Note

Notably, the step “export TKMROOT=$(pwd)” is necessary because the subpackage ‘’TinkerModeller.tkmtoolkit’’ is written in C++. This command is required for compiling the C++ code.

To build TinkerModeller, execute the following code in the terminal:

python -m build > build.log

This step will generate a folder called ‘’dist’’, which contains .whl file for environment setup:

pip dist/your_whl_name.whl

To make it straight, you could also use the shell script after getting into ‘’TinkerModeller’’ directory:

./reinstall

Testing After Installation

We always recommend that users run a test after every installation to ensure there are no bugs in the installation.

To automatically run the TinkerModeller tests, execute the following code in the terminal:

bash test.sh
pytest -n auto -s