Autonomous Synthesis Planner (ASP)

AI-Assisted Retrosynthetic Planning Framework

Autonomous Synthesis Planner (ASP) is an open-source software framework for generating, evaluating, and visualizing chemical synthesis pathways.

ASP provides a modular foundation for computational retrosynthesis workflows by combining:

• Molecular representation
• Reaction template libraries
• Retrosynthetic search
• Route scoring
• Synthesis pathway visualization

The goal of ASP is to accelerate chemical discovery by making synthesis planning workflows accessible, reproducible, and extensible.

---

Features

Current MVP Features

✅ SMILES-based molecular input

✅ Reaction template management

✅ Retrosynthetic planning engine

✅ Candidate route generation

✅ Route scoring framework

✅ JSON result export

✅ Command-line interface

✅ Python API

✅ Synthesis pathway visualization

---

Architecture

                 User
                  |
          +---------------+
          |  ASP CLI/API  |
          +---------------+
                  |
                  v
          +---------------+
          |    Planner    |
          +---------------+
                  |
       +----------+----------+
       |          |          |
       v          v          v

 Retrosynthesis  Scoring  Visualization

       |
       v

 Reaction Templates

       |
       v

 Molecule Representation

---

Installation

From source

git clone https://github.com/ksaad20/autonomous-synthesis-planner.git

cd autonomous-synthesis-planner

pip install -e .

---

Quick Start

Python API

from asp import ASP

planner = ASP()

result = planner.plan(
    "CC(=O)OC1=CC=CC=C1C(=O)O"
)

print(
    result
)

---

Command Line

Plan a molecule:

asp plan molecule "CCO"

Export results:

asp plan molecule "CCO" --output ethanol.json

Check installation:

asp version

---

Examples

ASP includes example workflows:

python examples/aspirin.py

python examples/caffeine.py

python examples/ibuprofen.py

Example targets:

Molecule	Application
Aspirin	Pharmaceutical synthesis
Caffeine	Heterocyclic chemistry
Ibuprofen	Drug molecule planning

---

Dataset Workflow

Prepare reaction data:

python scripts/download_data.py

Build reaction templates:

python scripts/build_templates.py

Generated templates are stored in:

datasets/templates/

---

Project Structure

autonomous-synthesis-planner/

├── examples/
├── datasets/
├── models/
├── scripts/
├── tests/
└── src/
    └── asp/
        ├── chemistry/
        ├── planning/
        ├── data/
        ├── visualization/
        ├── io/
        └── utils/

---

Development

Install development dependencies:

pip install -r requirements.txt

Run tests:

pytest

Run formatting checks:

black .
ruff check .

---

Scientific Scope

ASP is currently a research software MVP.

The current version focuses on:

• Software architecture
• Reproducible synthesis planning workflows
• Template-based retrosynthesis
• Extensible planning algorithms

Future versions aim to integrate:

• Machine learning reaction prediction
• Neural template extraction
• Cost-aware synthesis optimization
• Yield prediction
• Laboratory automation interfaces

---

Roadmap

v0.1.0 — MVP

• Core planning framework
• CLI
• Python API
• Template system
• Route scoring
• Visualization

v0.2.0

• Expanded reaction databases
• Improved search algorithms
• ML-based ranking

v1.0.0

• Autonomous synthesis planning platform
• Advanced AI models
• Experimental workflow integration

---

Contributing

Contributions are welcome.

Areas of interest:

• Chemistry algorithms
• Machine learning
• Molecular representation
• Optimization methods
• Scientific visualization

---

License

ASP is released under the Apache License 2.0.

See:

LICENSE

for details.

---

Citation

If you use ASP in research, please cite the repository release.

Autonomous Synthesis Planner (ASP)
Open-source retrosynthetic planning framework

---

Acknowledgements

ASP builds on decades of progress in:

• Computational chemistry
• Retrosynthesis
• Molecular informatics
• Artificial intelligence
• Scientific software engineering