The ability of artificial intelligence to predict protein folding is a real game changer in structural biology, but the predictions have several shortcomings. NKI researchers use their own algorithm to overcome some of these shortcomings.

Determining protein structures required labour-intensive analysis in the laboratory until the advent of AlphaFold and RoseTTAfold in 2021. Both methods use artificial intelligence to predict protein structures based on amino acid sequence. The development has been hugely influential, but the models lack biological context and interpretation. AlphaFold predicts only one state, even though proteins are highly dynamic. In addition, the protein models lack ligands. For example, haemoglobin needs haem to adopt its structure, but haem is not present in the AlphaFold model. These shortcomings inspired researchers in the group led by Anastassis (‘Tassos’) Perrakis at the Netherlands Cancer Institute (NKI) to combine databases and add more information to the model.