MOF Generation on HPC

Create new MOFs by combining generative AI and simulation on HPC.

Installation

The requirements for this project are defined using Anaconda.

Install the environment file appropriate for your system with a command similar to:

conda env create --file envs/environment-cpu.yml --force

If solving is slow try updating to the newest version of conda and using the libmamba solver:

conda update -n base conda
conda install -n base conda-libmamba-solver
conda config --set solver libmamba
conda env create --file envs/environment-cpu.yml

Running MOFA

The run_serial_workflow.py script defines a workflow using MOFA.

Set up the required input files by running 0_assemble-inputs.ipynb in input_files/zn-paddle-pillar.

Then invoke the workflow by running example-run.sh

The code will produce a run directory in run named using the start time and a hash of the run parameters.

The run directory contains the following files:

• run.log: The log messages produced during execution
• params.json: The arguments provided to the run script
• all-ligands.csv: A CSV file with the geometries of the generated ligands in XYZ format, if they passed all validation screens, and the SMILES string (if available).