MIT CSAIL researchers created FrameDiff, a computational tool utilizing machine learning to design novel protein structures. By simulating protein backbones with mathematical frames, FrameDiff constructs proteins that surpass natural varieties.

To advance our capabilities in protein engineering, MIT CSAIL researchers came up with “FrameDiff,” a computational tool for creating new protein structures beyond what nature has produced. The machine learning approach generates “frames” that align with the inherent properties of protein structures, enabling it to construct novel proteins independently of preexisting designs, facilitating unprecedented protein structures.

Journal Link

SE(3) diffusion model with application to protein backbone generation

Github Link

SE(3) diffusion model with application to protein backbone generation