Title: Proteins under pressure: Rethinking folding pathways through big data and atomistic modelling
Abstract:
Understanding protein folding in vivo demands models that go beyond idealized, dilute conditions. The cellular environment is crowded, complex, and dynamically structured by intricate water networks. In this study, we employ large-scale atomistic simulations, powered by newly developed big data and cloud-based algorithms, to reveal how molecular crowding and water network restructuring reshape the protein folding landscape. Our analyses show that crowding compacts unfolded ensembles and shifts folding pathways toward native-like conformations through entropic stabilization. At the same time, confined water networks emerge as active modulators, reorganizing hydration shells and altering kinetic barriers. Water is not a passive medium, it acts as a critical architect of the folding process. By leveraging high-throughput, cloud-enabled computation, we capture previously inaccessible intermediate states and dynamic fluctuations across large conformational ensembles. These results challenge classical folding paradigms and offer a new, data-driven perspective on how proteins navigate their rugged energy landscapes in biologically relevant environments. Our approach paves the way for more predictive models of proteostasis, misfolding diseases, and the design of next-generation biomaterials under realistic conditions.
