Abstract
Artificial intelligence (AI) has revolutionized structural biology but must be applied reliably. Here, we present an approach for derivation of distance restraints from AlphaFold structure predictions to aid in automated nuclear Overhauser effect (NOE) assignment during solution NMR structure determination. We describe steps for selecting reliable AlphaFold structure predictions, determination of atom distances, and generation of high-confidence distance restraints. This protocol can expedite solution NMR structure determination, reduce NOE assignment ambiguity, enhance accuracy, enable elucidation of elusive proteins, and validate structural predictions. For complete details on the use and execution of this protocol, please refer to Lin et al.(1).