min read

ACEGEN: A Reinforcement Learning Toolkit for Generative Drug Design

Published on
May 15, 2024

The advent of advanced machine learning techniques has opened new frontiers. Among these, reinforcement learning (RL) stands out for its potential to optimize and propose novel molecules with desired properties. Enter ACEGEN, a state-of-the-art toolkit designed to harness the power of RL for generative chemical design. Developed by our team at Acellera Therapeutics and collaborators, ACEGEN offers a comprehensive, flexible, and efficient platform tailored specifically for drug discovery.

The Need for ACEGEN

Drug design is a complex and multifaceted process involving the identification of biomolecules that exhibit optimal properties such as potency, selectivity, bioavailability, and toxicity. Traditional methods often fall short in efficiently navigating the vast chemical space, which contains an astronomical number of potential compounds. Reinforcement learning, with its trial-and-error approach and adaptive learning capabilities​​, presents a promising solution to this challenge.

What Makes ACEGEN Unique?

ACEGEN leverages TorchRL, a robust decision-making library within the PyTorch ecosystem. This integration ensures ACEGEN benefits from well-tested, modular components that enhance efficiency and reliability. The toolkit provides pre-trained models for various architectures, including GRU, LSTM, GPT-2, Llama2, and Mamba, which can be seamlessly integrated with custom architectures​​.

Key Features of ACEGEN

1. Modular and Flexible Design

ACEGEN is flexible, allowing researchers to incorporate custom scoring functions and models. The toolkit supports various molecular grammars, such as SMILES, DeepSMILES, SELFIES, and AtomInSmiles, providing a versatile platform for different molecular representations​​.

2. Comprehensive RL Algorithms

The toolkit includes implementations of several RL algorithms, such as REINFORCE, REINVENT, AHC, A2C, PPO, and PPOD. These algorithms are fully configurable, enabling customization to meet specific research needs. Notably, ACEGEN's implementation of these algorithms has been shown to achieve better results with faster training compared to existing solutions​​.

3. Efficient Training and Benchmarking

ACEGEN excels in efficiency, particularly in training and benchmarking tasks. For instance, ACEGEN significantly improved speed and performance metrics in comparisons between REINVENT implementations. The toolkit also offers a detailed tutorial for users to integrate custom scoring functions and models​​.

Benchmarking Performance

To assess the performance of different RL algorithms, ACEGEN was benchmarked using the Practical Molecular Optimization (MolOpt) framework. The results indicate that algorithms implemented in ACEGEN, such as PPOD and REINVENT-MolOpt, achieve state-of-the-art sample efficiency and performance in identifying desirable molecules. This benchmarking underscores ACEGEN's potential to significantly accelerate the drug discovery process by optimizing molecular generation more effectively than traditional methods​​.

Looking Ahead

ACEGEN represents a significant advancement in the field of generative drug design. By providing a robust, flexible, and efficient platform, ACEGEN empowers researchers to explore the vast chemical space more effectively, ultimately leading to the discovery of novel and more effective drugs. The open-source nature of ACEGEN, available under the MIT license, further ensures that the wider scientific community can benefit from and contribute to its ongoing development.

For more information and to access the ACEGEN toolkit, visit the ACEGEN GitHub repository.

Call for collaborations

We want to hear from you if you’re a pharma or biotech company looking to leverage generative AI on your chemical libraries or synthesis rules. We’re looking to partner with small and large companies to make ACEGEN an even more powerful solution for drug discovery.

If you’re interested, get in touch.


  1. Bou, A., Thomas, M., Dittert, S., et al. "ACEGEN: Reinforcement Learning of Generative Chemical Agents for Drug Discovery." GEM workshop, ICLR 2024.
  2. Blaschke, T., Arús-Pous, J., Chen, H., et al. "REINVENT 2.0: An AI Tool for De Novo Drug Design." Journal of Chemical Information and Modeling, 60(12):5918-5922, 2020.
  3. Gao, W., Fu, T., Sun, J., et al. "Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization." Advances in Neural Information Processing Systems, 35:21342-21357, 2022.

Occasional newsletter
Receive important news, updates, tutorials, and the latest by the Acellera team.
Read about our .
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.