Accurate conformational ensembles of intrinsically disordered proteins using reweighting based on NMR chemical shifts

Intrinsically disordered proteins and protein regions (IDRs) underpin a wide range of vital biological processes but exhibit dynamic and heterogeneous conformations. Currently, many computational efforts seek to elucidate the conformational ensembles of these disordered proteins, yet most methods still struggle to fully capture their structural diversity. Here, we integrate structural libraries of various IDRs-derived from coarse-grained molecular dynamics (MD) simulations and machine learning models-with experimental chemical shifts obtained from NMR spectroscopy. Through a maximum entropy reweighting approach, we obtain reliable ensembles that more accurately reflect observed chemical shifts and reveal transient states. Our results highlight the importance of comprehensive sampling strategies for capturing diverse conformational states. Furthermore, we show that these weighted ensembles faithfully track conformational rearrangements under various conditions such as temperature, mutational effects, and environment, which are not fully captured by experiments alone. This approach provides a dataset encompassing each IDR's specific structures along with their weights, offering a foundation for systematically exploring IDR structural landscapes, refining our understanding of their functional roles, and shedding light on processes related to misfolding and aggregation.