Can AI predict the 3d structure of any protein from its amino acid sequence ?

AlphaFold 2 solved a 50-year grand challenge in biology with near-experimental accuracy at CASP14. It now powers most structural biology pipelines.

Currently, AI can predict the 3D structure of many proteins from their amino acid sequence with a high degree of accuracy, thanks to advances in deep learning algorithms and large datasets of known protein structures. The AlphaFold model, developed by DeepMind, is a notable example of such an AI system, which has achieved state-of-the-art performance in protein structure prediction. However, predicting the structure of any protein remains a challenging task, particularly for those with complex or unusual folds, and ongoing research is focused on improving the accuracy and robustness of these models. While significant progress has been made, there are still limitations and uncertainties in protein structure prediction, especially for certain types of proteins or sequences.

— Enriched May 9, 2026 · Source: Nature — https://www.nature.com