Science Score: 67.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
    Found 4 DOI reference(s) in README
  • Academic publication links
    Links to: arxiv.org, zenodo.org
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (14.8%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

Basic Info
  • Host: GitHub
  • Owner: schwallergroup
  • License: mit
  • Language: Python
  • Default Branch: main
  • Size: 10.3 MB
Statistics
  • Stars: 37
  • Watchers: 0
  • Forks: 2
  • Open Issues: 2
  • Releases: 0
Created over 1 year ago · Last pushed 8 months ago
Metadata Files
Readme Contributing License Code of conduct Citation

README.md

Chemical Reasoning in LLMs for Synthesis Planning and Mechanism Elucidation

tests DOI:10.48550/arXiv.2503.08537 PyPI PyPI - Python Version Documentation Status Code style: black

Overview of LLMs as chemical reasoning engines

Overview

This repository contains the implementation of a novel framework that leverages LLMs as chemical reasoning engines to guide traditional search algorithms in chemistry. Our approach demonstrates how LLMs can be effectively used for:

  1. Strategy-aware Retrosynthetic Planning: Enable chemists to specify desired synthetic strategies in natural language and find routes that satisfy these constraints.
  2. Mechanism Elucidation: Guide the search for plausible reaction mechanisms by combining chemical principles with systematic exploration.

Key Features

  • 🧪 Natural language specification of synthetic strategies
  • 🔍 LLM-guided search through chemical space
  • 📊 Benchmark datasets for both synthesis planning and mechanism elucidation
  • 🤖 Support for multiple LLM providers (Claude, GPT-4, DeepSeek)

Installation

```bash

Install from PyPI (TBD)

pip install steer

Install from source

pip install git+https://github.com/schwallergroup/steer.git ```

Quick Start

Steerable Synthesis Planning

```bash

Run the complete synthesis benchmark

steer synth --model=claude-3-5-sonnet bench

Run a single task

steer synth --model=claude-3-5-sonnet bench --task=ea8df340d54596eda93e23f04dff3a9b ```

Mechanism Finding

```bash

Run mechanism elucidation benchmark

steer mech --model=claude-3-5-sonnet bench ```

Benchmarks

The repository includes two main benchmarks:

Synthesis Planning Benchmark

  • Multiple target molecules of varying complexity
  • Strategic constraints specified in natural language
  • Evaluation metrics for route-to-prompt alignment

Mechanism Elucidation Benchmark

  • 12 diverse organic reactions
  • Ground truth mechanisms with elementary steps
  • Performance metrics for mechanism prediction

Citation

If you use this work in your research, please cite:

bibtex @misc{bran2025chemicalreasoningllmsunlocks, title={Chemical reasoning in LLMs unlocks steerable synthesis planning and reaction mechanism elucidation}, author={Andres M Bran and Theo A Neukomm and Daniel P Armstrong and Zlatko Jončev and Philippe Schwaller}, year={2025}, eprint={2503.08537}, archivePrefix={arXiv}, primaryClass={cs.AI}, url={https://arxiv.org/abs/2503.08537}, }

Development

Click to expand development instructions ### Setup Development Environment ```bash git clone https://github.com/schwallergroup/steer.git cd steer pip install -e . ``` ### Running Tests ```bash pip install tox tox ```

License

MIT License

Contributors

  • Andres M Bran
  • Théo A. Neukomm
  • Daniel Armstrong
  • Zlatko Jončev
  • Philippe Schwaller

Contact

For questions and feedback: - 📧 Email: andres.marulandabran@epfl.ch, philippe.schwaller@epfl.ch - 🌐 Schwaller Group Website

Owner

  • Name: schwallergroup
  • Login: schwallergroup
  • Kind: organization

Citation (CITATION.cff)

cff-version: 1.0.2
message: "If you use this software, please cite it as below."
title: "steer"
authors:
  - name: "Andres M Bran"
version: 0.0.1
doi:
url: "https://github.com/schwallergroup/steer"

GitHub Events

Total
  • Create event: 5
  • Issues event: 4
  • Watch event: 36
  • Delete event: 1
  • Issue comment event: 4
  • Member event: 2
  • Public event: 1
  • Push event: 28
  • Pull request review comment event: 1
  • Pull request review event: 1
  • Pull request event: 18
  • Fork event: 2
Last Year
  • Create event: 5
  • Issues event: 4
  • Watch event: 36
  • Delete event: 1
  • Issue comment event: 4
  • Member event: 2
  • Public event: 1
  • Push event: 28
  • Pull request review comment event: 1
  • Pull request review event: 1
  • Pull request event: 18
  • Fork event: 2

Dependencies

.github/workflows/tests.yml actions
  • actions/checkout v2 composite
  • actions/setup-python v2 composite
  • codecov/codecov-action v1 composite
pyproject.toml pypi