Diffusion-based Molecule Generation with Informative Prior Bridges
This work addresses the problem of generating high-quality molecules for biomedical applications like antibody design, offering an incremental improvement over existing diffusion-based methods.
The authors tackled the challenge of generating realistic 3D molecules by incorporating physical and statistical priors into diffusion models, resulting in improved quality and stability scores for molecule structures and more uniformly distributed point clouds.
AI-based molecule generation provides a promising approach to a large area of biomedical sciences and engineering, such as antibody design, hydrolase engineering, or vaccine development. Because the molecules are governed by physical laws, a key challenge is to incorporate prior information into the training procedure to generate high-quality and realistic molecules. We propose a simple and novel approach to steer the training of diffusion-based generative models with physical and statistics prior information. This is achieved by constructing physically informed diffusion bridges, stochastic processes that guarantee to yield a given observation at the fixed terminal time. We develop a Lyapunov function based method to construct and determine bridges, and propose a number of proposals of informative prior bridges for both high-quality molecule generation and uniformity-promoted 3D point cloud generation. With comprehensive experiments, we show that our method provides a powerful approach to the 3D generation task, yielding molecule structures with better quality and stability scores and more uniformly distributed point clouds of high qualities.