Molecular design
FreeList of molecular design using Generative AI and Deep Learning.
FreeFree tier
About Molecular design
A curated, community-maintained list of research papers focused on molecular and material design using generative AI and deep learning techniques. Hosted on GitHub under the GPL-3.0 license, this repository organizes publications by method (e.g., GANs, VAEs, diffusion models, transformers, GNNs, reinforcement learning) and application (drug design, material design). It serves as a comprehensive starting point for researchers and practitioners looking to explore state-of-the-art deep learning approaches for molecular generation and optimization.
Key Features
Covers multiple deep learning paradigms: GANs, VAEs, diffusion models, transformers, GNNs, RNNs, reinforcement learning, energy-based models, score-based models, and prompt learning
Organized by algorithm type and application area (drug design, material design)
Open-source repository with 946 stars and active community contributions
Includes molecular optimization and property prediction resources
Licensed under GPL-3.0 for free use and redistribution
Pros & Cons
Pros
- Comprehensive collection of 1,000+ papers spanning diverse methods and applications
- Actively maintained with recent commits (1,009 commits on main branch)
- Well-tagged with relevant topics for easy filtering (e.g., ‘drug-design’, ‘diffusion’, ‘vae’)
- Free and open source with permissive GPL-3.0 license
- Useful for both beginners and experts to quickly survey the field
Cons
- Not a software tool – requires manual reading of papers to apply methods
- Papers are not ranked or quality-filtered beyond inclusion
- No direct implementation code provided; only references to external papers
- Limited to deep learning approaches; excludes classical computational chemistry methods
Best For
Research literature review in AI-driven drug discovery and materials scienceEducational resource for students and researchers learning deep generative models in chemistryBenchmarking and comparison of generative modeling methods for molecular designIdentifying state-of-the-art approaches for molecular optimization tasks
FAQ
What types of deep learning methods are covered in this repository?
The repository covers GANs, VAEs, diffusion models, transformers, RNNs, GNNs, reinforcement learning, energy-based models, score-based models, and prompt learning, among others.
Is this a software tool or a paper list?
This is a curated list of research papers, not a software tool. Each entry links to a paper (typically on arXiv or publisher site).
How can I contribute to this repository?
You can fork the repository and submit pull requests with new papers or corrections, following the contribution guidelines on GitHub.
What license does this repository use?
The repository is licensed under the GPL-3.0 license.