DNA Diffusion

This project investigates the application and adaptation of recent diffusion models to genomics data. Diffusion models are powerful models with outstanding results for image generation (e.g., Stable Diffusion, DALL-E) and music generation (recent version of the magenta project). A particular model formulation called "guided" diffusion allows biasing the generative process toward a specific direction if, during training, a text or continuous/discrete labels are provided. This enables the creation of "AI artists" that, based on a text prompt, can create beautiful and complex images.