☁️ Cloud-Bind

Cloud-Based Drug Binding Structure Prediction

---

🔬 Overview

Cloud-Bind is a cloud-powered platform for running molecular docking, virtual screening, and ligand pose evaluation directly in Google Colab, no local installation or high-performance computing required.

This repository contains a collection of Jupyter Notebooks integrating:

• GNINA: Molecular docking with deep learning scoring using convolutional neural networks (CNNs).
• Uni-Dock: GPU-accelerated molecular docking for ultralarge virtual screening.
• OpenBPMD: Binding pose metadynamics for ligand stability evaluation.

The main goal of Cloud-Bind is to demonstrate how cloud computing can democratize access to advanced drug discovery tools, providing a low-cost and fully reproducible pipeline for structure-based modeling and binding pose assessment.

---

🚀 Available Workflows

GNINA

1. GNINA
   Perform molecular docking with integrated CNN-based scoring and optimization.

2. GNINA + MD
   Combine GNINA docking with OpenMM molecular dynamics simulations for pose relaxation and stability.

3. GNINA + OpenBPMD
   Use GNINA docking and OpenBPMD metadynamics to evaluate ligand pose stability (BPMD).

---

Uni-Dock

4. Uni-Dock
   Perform GPU-accelerated molecular docking with multiple scoring functions (vina, vinardo, etc.).

5. Uni-Dock + MD
   Combine Uni-Dock docking with OpenMM molecular dynamics simulations for post-docking relaxation.

6. Uni-Dock + OpenBPMD
   Integrate Uni-Dock docking with OpenBPMD to assess binding pose stability using metadynamics.

---

Virtual Screening

7. Virtual Screening
   Execute an end-to-end virtual screening protocol that includes:
   • Deep learning docking with GNINA
   • Conformational ensemble generation via PLACER
   • Binding affinity prediction with AEV-PLIG

---

🐞 Reporting Issues

If you encounter any bugs or have suggestions, please open an issue at:
👉 https://github.com/pablo-arantes/Cloud-Bind/issues

---

🙏 Acknowledgments

We gratefully acknowledge the developers of the tools integrated in Cloud-Bind:

• GNINA — Deep learning–based molecular docking
• Uni-Dock — GPU-accelerated docking for ultralarge virtual screening
• OpenBPMD — Binding pose metadynamics
• ProLIF — Protein–Ligand Interaction Fingerprints
• py3Dmol — Interactive 3D molecular visualization

Special thanks to:
Pablo R. Arantes, Conrado Pedebos, and Rodrigo Ligabue-Braun, developers of Cloud-Bind
and to the Making it Rain team for pioneering open, cloud-based molecular simulation workflows.

---

📖 Citation Guidelines

If you use Cloud-Bind or any of its integrated notebooks, please cite the respective software and references:

• Cloud-Bind
  Pedebos et al., ChemRxiv, 2025, DOI: 10.26434/chemrxiv-2025-jr04j

• GNINA
  McNutt et al., J. Cheminform. 2021, DOI: 10.1186/s13321-021-00522-2

• Uni-Dock
  Yu et al., J. Chem. Theory Comput. 2023, DOI: 10.1021/acs.jctc.2c01145

• Molecular Dynamics Notebooks
  Arantes et al., J. Chem. Inf. Model. 2021, DOI: 10.1021/acs.jcim.1c00998

• OpenBPMD
  Lukauskis et al., J. Chem. Inf. Model. 2022, DOI: 10.1021/acs.jcim.2c01142

---

💡 Cloud-Bind: Bringing advanced molecular docking and pose evaluation to everyone, powered by the cloud.