From Handbooks to High-Throughput: Automated application of Woodward-Fieser, Fieser and Fieser-Kuhn rules to predict low-energy ππ^∗ absorption maxima of α,β-unsaturated carbonyl compounds, dienes and systems with more than four conjugated carbon-carbon double bonds. Application of extended Woodward-Fieser rules to predict the absorption maxima of 3,4,6-substituted coumarins.

Woodward-Type Examples

CC(=O)C1=C(C)CCCC1 Experiment: 247 nm Woodward: 237 nm Woodward refined: 246 nm

CC1=C2CCCCC2CCC1=O Experiment: 239 nm Woodward: 242 nm Woodward refined: 251 nm

CC(=O)C=CC1=C(C)CCCC1(C)C Experiment: 283 nm Woodward: 281 nm Woodward refined: 281 nm

---

Installation

This project is based on RDKit and the tutorials are provided as Jupyter notebooks. All dependencies are listed in the requirements.txt file. To install them the pyproject.toml can be used via:

pip install -e

Alternatively, you can install the packages directly:

pip install numpy pandas rdkit seaborn jupyterlab notebook py3Dmol

Usage

The basic usage of the tool is outlined in the . For example, to predict the absorption maximum of the enon shown above, one can run:

import chromopredict as cp

# original woodward-fieser rules
abs_max, description, image = cp.predict(
  smiles='CC(=O)C1=C(C)CCCC1',
  solvent=None,
  verbose=True, # return increments of all structural features
  chromlib='woodward')

#refined rules by us
abs_max, description, image = cp.predict(
  smiles='CC(=O)C1=C(C)CCCC1',
  solvent=None,
  verbose=True,
  chromlib='woodward_refine')

Citation