Sequence Prediction, Structural Modelling and the Challenges of Plasticity in Coiled Coils

The coiled coil is one of the most widespread and versatile protein folding motifs and is involved in a vast array of biological functions. The study and design of these assemblies have been shaped by the development of specialised computational tools. This review charts the landscape of these in silico methods, beginning with the foundational sequence-based algorithms for coiled-coil prediction and classification, before moving to the powerful parametric methods that enable the generation of idealised, atomic-resolution structural models. We argue that while indispensable, these tools can promote a view of coiled coils as static, singular structures. Challenging this, we highlight the structural plasticity of coiled coils, the remarkable and increasingly evident ability of these assemblies to adopt multiple distinct conformations and dynamically switch between states. To understand the energetic and mechanistic principles governing this complex behaviour, we explore the crucial role of molecular dynamics (MD) simulations in providing atomistic insights that are inaccessible to static modelling. Finally, we look forward, considering how the next generation of coiled-coil bioinformatics must evolve to address these challenges to take advantage of the opportunities presented by the current explosion in genomic sequence data and the proliferation of AI-predicted protein structures.