Deciphering Oxime Biocatalysis (A QM/MM Tutorial)

Author:

Amit Singh Sahrawat, University of Graz (amit.amit@uni-graz.at)

Tutorial:

1.0

Date:

Apr 20, 2026

Note

This project is under active development.

Warning

Please raise any problems in the issue tracker.

Objective

To decipher the newly discovered mechanistic intricacies of the conversion of oximes to amines via imines by reductive dehydration. Using QM/MM simulations, we aim to model the molecular structures corresponding to the respective reaction coordinates (Reactant, Transition States and Products) for each reaction pathway. Finally, the energy barrier is computed using QM/MM Steered Molecular Dynamics (SMD) simulations. For full details of the science behind this tutorial, please read Sahrawat et al. (2024):

Sahrawat, A. S. Deciphering the Unconventional Reduction of C=N Bonds by Old Yellow Enzymes Using QM/MM. ACS Catalysis 2024. doi:10.1021/acscatal.3c04362

Tutorial files

All of the necessary tutorial files can be found on GitHub in the hopanoid/Enzyme-Reaction-Dynamics-Tutorial directory, which can be easily obtained by git-cloning the repository:

git clone https://github.com/hopanoid/Enzyme-Reaction-Dynamics-Tutorial.git

How to Cite

If this tutorial contributes to your research, please cite the accompanying peer-reviewed article:

Citation

Sahrawat, A. S. Deciphering the Unconventional Reduction of C=N Bonds by Old Yellow Enzymes Using QM/MM. ACS Catalysis 2024. doi:10.1021/acscatal.3c04362

A CITATION.CFF file is included in the repository for automated citation tools (e.g., GitHub’s Cite this repository button).

Workflow overview

For this tutorial we’ll use Amber (v22) to set up the system, combined with TeraChem (v1.96H-beta, CUDA 12.1) to run the QM/MM simulations, and NBO (v7.0) to perform orbital analysis. Exact version details and hardware requirements are documented in the Hardware Requirements page. An initial structure is provided, which can be found in the tutorial/metadata/input_structures directory, as well as the input files that are necessary for running Amber. The overall workflow consists of the following steps:

Contents

• Protein Parametrisation with AMBER
  • Preparing Your Enzyme Structure for AMBER
    • Splitting the PDB in protein, ligand and water molecules
    • Assessing the protonation states of the enzyme with careful attention for active site residues
    • Setting Histidine Protonation State with AMBER LEaP
• Ligand Parametrisation with Antechamber
  • Know Your Ligand: pKa and Ionisation State
    • Generating point charges and assigning atomtypes with antechamber
    • Looking for missing force field parameters with parmchk2
• System Setup with tLEaP
  • Assembling the Enzyme-Ligand-Solvent System
    • Building the Solvated System with tLEaP
• System Equilibration and Minimisation
  • Removing bad atomic contacts and minimizing vacuum!
• QM Region Selection and Charge Analysis
  • Selecting Active-Site Residues for the QM Region
• QM/MM Production Run with TeraChem
• QM/MM Geometry Optimisation with Qsite
  • Chose rightly and truncate wisely!
• QM/MM Steered Molecular Dynamics (SMD)
  • Defining the Collective Variable (CV) for Hydride Transfer
• Natural Bond Orbital (NBO) Analysis
  • NBO Coupled with QM/MM SMD Simulations
• Adapting This Tutorial to Your System
  • 1. Input structure
  • 2. Residue and ligand names
  • 3. Protonation state residue numbers
  • 4. QM region atom masks
  • 5. QM total charge
  • 6. Collective variable atom indices
  • 7. Number and length of simulations
• Hardware Requirements
  • Software Versions Used in This Tutorial
  • What You Need to Run This Tutorial
    • TeraChem: GPU is Essential, CUDA Cores are the Key
    • AMBER pmemd.cuda: GPU-Accelerated Classical MD
    • AMBER sander: QM/MM Runs on CPU
    • NBO: CPU-Only, Scales with Core Count
    • Summary: a Practical Single-Workstation Setup
• Glossary
• Alternative QM/MM Tutorials
• Scripts Reference
  • Pre-processing
  • Simulations