QuantumBind-RBFE

Summary

QuantumBind-RBFE is Acellera's implementation for Relative Binding Free Energy (RBFE) calculations. This approach is based on the Alchemical Transfer Method (ATM) and incorporates Neural Network Potentials (NNP) within a hybrid NNP/MM framework. The data in this repository was used in the accompanying manuscript: QuantumBind-RBFE: Accurate Relative Binding Free Energy Calculations Using Neural Network Potentials.

Contents

• inputs/: Contains input files for binding free energy calculations.
• results/: Includes ΔΔG and ΔG values from the runs presented in our manuscript.
• tutorial/: Provides a Jupyter notebook offering a step-by-step tutorial on setting up and running a QuantumBind-RBFE calculation.
• scripts/: Contains the necessary scripts to execute the production phase using the provided input files.

Prerequisites

QuantumBind-RBFE can be run via HTMD, Acellera's molecular modeling environment. Ensure that you have HTMD installed and properly configured to execute the provided scripts and workflows.

References

• AceForce:
  If you wish to use the AceForce model, Acellera's own NNP, it can be found on Hugging Face: AceForce 1.0.
  AceForce 1.0 is available for non-profit use and demonstrations, allowing researchers to explore its potential in their projects.