Abstract
Small-angle X-ray fiber diffraction has informed much of what we know regarding the molecular events during muscle contraction but robust tools for predicting X-ray fiber patterns from muscle have been lacking. A complication in formulating such tools is the dynamic, stochastic nature of the sarcomere structures during contraction where individual myofilaments undergo deformations due to nonuniform strain generated by the myosin crossbridges. Here, we address this need with a "forward problem" approach using a spatially explicit model (MUSICO) to predict the molecular configurations responsible for the observed muscle force and use these configurations to predict the diffraction patterns that can be compared to experiments. We combine this with a newly developed, rigorous formulation, presented here, for the calculation of 2D diffraction patterns from actin filaments under nonuniform strain. We compare all-atom predictions to coarse-grained simulations to show how much information is lost by coarse-graining, and discuss the results in the context of diffraction patterns currently obtainable experimentally. We show that most low-resolution coarse-grained models in the literature suffice for prediction of meridional peak shapes for the purposes of estimating force distributions in the actin filaments, but accurate prediction of layer line intensities require much higher resolution models, including the all-atom models as presented here. These developments represent an important step towards our long-term goal of using molecular simulations to interpret X-ray fiber diffraction patterns from striated muscle during active contraction.