pdb2reaction Documentation

Version: v0.3.7


pdb2reaction workflow overview

pdb2reaction is a Python CLI toolkit for automated enzymatic reaction-path modeling directly from PDB structures using machine-learning interatomic potentials (MLIPs).


Start here

Goal

Workflow

First end-to-end run

Quickstart: all

Reactant only

Quickstart: scan

TS candidate available

Quickstart: tsopt + freq

Run failure / error

Common Error Recipes

Refer to Installation for prerequisites.


Subcommands

Subcommand

Description

all

End-to-end workflow: extraction → scan → MEP → TS optimization → IRC → thermochemistry → DFT

extract

Extract active site model (binding pocket) from protein–ligand complex

fix-altloc

Resolve PDB alternate locations

add-elem-info

Repair PDB element columns (77–78)

opt

Single-structure geometry optimization (L-BFGS or RFO. [+ optional flatten])

tsopt

Transition state optimization (Dimer or RS-I-RFO. [+ optional flatten])

path-opt

Single-step MEP optimization via GSM or DMF (from 2 structures)

path-search

Recursive multi-step MEP search with automatic refinement (2+ structures)

scan

1D bond-length driven scan with restraints

scan2d

2D distance grid scan

scan3d

3D distance grid scan

freq

Vibrational frequency analysis & thermochemistry

irc

Intrinsic Reaction Coordinate calculation

dft

Single-point DFT calculations (GPU4PySCF / PySCF)

trj2fig

Plot energy profiles from XYZ trajectories

energy-diagram

Draw an energy diagram from numeric values

bond-summary

Detect and report covalent bond changes between consecutive structures


Configuration & Reference

Topic

Page

Getting Started

Getting Started

CLI conventions & input requirements

CLI Conventions

Cluster boundary atoms (link hydrogens, --freeze-atoms)

Frozen Atoms

Common errors & fixes

Troubleshooting

CLI command reference

Command Reference

YAML configuration options

YAML Reference

MLIP backend settings

MLIP Calculator

Terminology

Glossary


System Requirements

Hardware

  • OS: Linux

  • GPU: CUDA 12.x compatible

  • VRAM: 8 GB+ recommended

  • RAM: 16 GB+ recommended

Software

  • Python >= 3.11

  • PyTorch with CUDA support

  • CUDA 12.x toolkit

For setup, see Installation.


Agent Skills

pdb2reaction ships AI-agent instructions under .claude/skills/ covering CLI subcommands, structure I/O, backend installation, workflows, output parsing, and HPC operation. See .claude/skills/README.md for the full skill index and installation instructions.


Citation

A preprint describing pdb2reaction is in preparation. Currently, if you find this work helpful for your research, please cite the software itself:

@software{ohmura2026pdb2reaction,
  author       = {Ohmura, Takuto},
  title        = {pdb2reaction},
  year         = {2026},
  month        = {4},
  version      = {0.3.6},
  url          = {https://github.com/t-0hmura/pdb2reaction},
  license      = {GPL-3.0},
  doi          = {10.5281/zenodo.19197865}
}

License

pdb2reaction is distributed under the GNU General Public License version 3 (GPL-3.0).


Getting Help

# General help
pdb2reaction --help

# Command help
pdb2reaction <subcommand> --help

# Advanced options (dry-run, internal tuning, etc.)
pdb2reaction <subcommand> --help-advanced

For issues and feature requests, visit the GitHub repository.