Sovereign 8s.is mirror of standardgalactic/legion-scaffold-decorator
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
Vlad Aladinskii 602ab3edd4 Initial commit
2026-01-05 18:49:11 +04:00
img Initial commit 2026-01-05 18:49:11 +04:00
input Initial commit 2026-01-05 18:49:11 +04:00
requirements Initial commit 2026-01-05 18:49:11 +04:00
src Initial commit 2026-01-05 18:49:11 +04:00
test Initial commit 2026-01-05 18:49:11 +04:00
.flake8 Initial commit 2026-01-05 18:49:11 +04:00
.gitignore Initial commit 2026-01-05 18:49:11 +04:00
.pre-commit-config.yaml Initial commit 2026-01-05 18:49:11 +04:00
LICENSE Initial commit 2026-01-05 18:49:11 +04:00
pyproject.toml Initial commit 2026-01-05 18:49:11 +04:00
README.md Initial commit 2026-01-05 18:49:11 +04:00

Scaffold Decorator for the LEGION Workflow

License: MIT Python 3.9.18 RDKit ChemRxiv

A Python package for generating molecules by combining molecular scaffolds with decorations.

The code and data to reproduce combinatorial generation of molecular structures from LEGION workflow.

Read the preprint: Molecular LEGION: Latent Enumeration, Generation, Integration, Optimization and Navigation. A case study of incalculably large chemical space coverage around the NLRP3 target

Image

🎯 Quick Start

Installation

# Create and activate environment
conda create -n scaffold_decorator python=3.9.18
conda activate scaffold_decorator

# Install RDKit
conda install -c conda-forge rdkit

# Install package and dependecies
pip install -e .

📁 Input files

All required input data are available in the input/ derictory:

  • unique_scaffolds.csv
  • unique_left_decorations.csv
  • unique_right_decorations.csv

🎲 Decorate scaffolds

# Random generation - N random molecules
generate-random -n 100

# Systematic generation - all combinations with sampling of 64 decorations in every position
generate

⚙️ Custom options

# Random generation with custom parameters
generate-random \
    --scaffolds-csv <path to scaffolds csv> \
    --left-decorations-csv <path to left decorations csv> \
    --right-decorations-csv <path to right decorations csv> \
    --n-molecules 100 \
    --output <path to output csv> \
    --no-randomize-scaffolds \
    --no-randomize-decorations

# Systematic generation with custom parameters
generate \
    --scaffolds-csv <path to scaffolds csv> \
    --left-decorations-csv <path to left decorations csv> \
    --right-decorations-csv <path to right decorations csv> \
    --output-dir <path to output directory> \
    --decorations-sample-size 10 \
    --save-every 1 \
    --no-shuffle

📋 Parameters

Random Generation (generate_random.py)

Parameter Description Default
--scaffolds-csv CSV file with scaffolds /input/unique_scaffolds.csv
--left-decorations-csv CSV file with left decorations /input/unique_left_decorations.csv
--right-decorations-csv CSV file with right decorations /input/unique_right_decorations.csv
-n, --n-molecules Number of molecules to generate required
--output Output CSV file output/random_molecules.csv
--no-randomize-scaffolds Use scaffolds systematically False
--no-randomize-decorations Use decorations systematically False

Systematic Generation (generate.py)

Parameter Description Default
--scaffolds-csv CSV file with scaffolds /input/unique_scaffolds.csv
--left-decorations-csv CSV file with left decorations /input/unique_left_decorations.csv
--right-decorations-csv CSV file with right decorations /input/unique_right_decorations.csv
--output-dir Output directory /output
--decorations-sample-size Number of decorations to sample 64
--save-every Save results every N scaffolds 10
--no-shuffle Don't shuffle scaffolds False

📋 Data Format

Input CSV files must contain a smiles column:

smiles
c1ccc([*:1])cc1
[*:1]c1ccc([*:2])cc1

Output CSV files contains:

smiles,scaffold,left_decoration,right_decoration
Cc1ccccc1O,c1ccc([*:1])cc1,[*:1]C,[*:2]O

License:

This project is licensed under the MIT License - see the LICENSE file for details.