Role of artificial intelligence in advancing cryo electron microscopy

Structural information available for large macromolecular complexes still remains only a small fraction of what exists in nature. Cryo-electron microscopy (cryo-EM) has greatly advanced our ability to visualise these complexes at near-atomic resolution, providing valuable insights into their dynamics and function. While improvements in sample preparation and instrumentation have led to a surge in solved structures, major challenges remain in data processing. Handling big and heterogeneous datasets with traditional computational methods is slow and resource-intensive. The rise of artificial intelligence (AI) and machine learning (ML) is reshaping this landscape by enabling more automated, efficient and economical approaches. Such tools are now being integrated across multiple stages of the cryo-EM workflow. In this chapter, we highlight these developments that hold the potential to accelerate structure determination and broaden our understanding of complex biomolecular systems.