Alchemd

A computational toolkit for alchemical free energy perturbation (FEP) calculations in computer-aided drug design

Overview

Alchemd is a Python-based toolkit for Relative Binding Free Energy (RBFE) calculations. The software provides an integrated platform for molecular preparation, dynamics simulation, and thermodynamic analysis with enhanced convergence algorithms.

System Requirements

• Operating System: Linux (Ubuntu 18.04+) or Windows with WSL2
• Hardware: NVIDIA GPU with CUDA support (RTX 4060 Ti+ recommended)
• Dependencies: CUDA 12.0, Conda/Mamba, GCC 9+
• License: Amber license required (apply here)
• Optional dependency: Gaussian (for NSAA non-standard amino acid module)

Installation

Automated Installation

Prerequisites: Before running the automated installation scripts, ensure you have:

• Conda or Mamba installed and properly configured
• CUDA 12.0 installed and accessible in your system PATH

Execute the following installation scripts sequentially:

# 1. System environment setup
chmod +x install/setup_linux.sh
./install/setup_linux.sh

# 2. Python environment configuration
chmod +x install/setup_conda.sh
./install/setup_conda.sh

# 3. PMEMD compilation (requires Amber license)
chmod +x install/build_amber.sh
./install/build_amber.sh /path/to/pmemd24

Installation Guide

For detailed installation procedures, refer to:

• docs/Installation_Guide_EN.md

Quick Start

# Structure preparation
python core/prepare_file.py -p protein.pdb -l ligands.sdf -o workspace --cs-align -d -auto

# Molecular dynamics simulation  
python core/run_md.py -c -ln 50 -p ./workspace/prepare -m Amber -auto

# Results analysis
python core/analyze_result.py -auto -c -w ./workspace/run -m Amber

# GUI interface
python gui/gui.py

Documentation

• User Guide: docs/Alchemd_User_Guide_EN.md
• Installation Guide: docs/Installation_Guide_EN.md
• Example Data: example/ directory

Citation

If you use Alchemd in your research, please cite:

@software{alchemd2025,
  title={Alchemd: A toolkit for alchemical free energy perturbation calculations},
  year={2025},
  url={https://github.com/ZheLi-Lab/Alchemd-toolkit}
}

Methodological References:

• CAR method: DOI
• cs-fep implementation: DOI

License

Alchemd is dual-licensed:

• Non-commercial use: Licensed under CC BY-NC-SA 4.0 - free for personal, academic, and non-profit use
• Commercial use: Requires separate commercial license - contact us for licensing inquiries

See the LICENSE file for complete terms and conditions.

Support

For questions and issues, please use GitHub Issues.