Aureka Releases OpenDDE Open Source Drug Discovery Engine

Aureka announced in a press release the release of Open Drug Discovery Engine (OpenDDE), an open source all atom biomolecular foundation model designed to support AI based therapeutic research. The system serves as a structural reasoning core for modern drug discovery workflows.

OpenDDE models interactions among proteins, nucleic acids, small molecule ligands, and other biomolecular elements through a biomolecular co folding approach. It is designed as a shared structural reasoning layer to connect sequence, structure, and function modeling. The system enables structure prediction and lays groundwork for design, affinity estimation, optimization, and closed loop discovery tasks.

The model includes about 655 million trainable parameters and required roughly 414,000 GPU hours to train, equivalent to about 54 years on one computing unit. Aureka is releasing the training code, inference pipelines, checkpoints, and benchmarks under the Apache 2.0 license to enable independent validation and community collaboration.

Across in silico benchmarks, OpenDDE shows competitive performance in co folding tasks and approaches results reported for IsoDDE level systems while maintaining an open and reproducible framework for the research community.