Extended Sampling\nof Macromolecular Conformations\nfrom Uniformly Distributed Points on Multidimensional Normal Mode\nHyperspheres

Proteins are dynamic entities that adopt diverse conformations,\nwhich play a pivotal role in their function. Understanding these conformations\nis essential, and protein collective motions, particularly those captured\nby normal mode (NM) and their linear combinations, provide a robust\nmeans for conformational sampling. This work introduces a novel approach\nto obtaining a uniformly oriented set of a given number of lowest\nfrequency NM combined vectors and generating harmonically equidistant\nrestrained structures along them. They are all thus uniformly located\non concentric hyperspheres, systematically covering the defined NM\nspace fully. The generated structures are further relaxed with standard\nmolecular dynamics (MD) simulations to explore the conformational\nspace. The efficiency of the approach we termed “distributed\npoints Molecular Dynamics using Normal Modes” (dpMDNM) was\nassessed by applying it to hen egg-white lysozyme and human cytochrome\nP450 3A4 (CYP3A4). To this purpose, we compared our new approach with\nother methods and analyzed the sampling of existing experimental structures.\nOur results demonstrate the efficacy of dpMDNM in extensive conformational\nsampling, particularly as more NMs are considered. Ensembles generated\nby dpMDNM exhibited a broad coverage of the experimental structures,\nproviding valuable insights into the functional aspects of lysozyme\nand CYP3A4. Furthermore, dpMDNM also covered lysozyme structures with\nrelatively elevated energies corresponding to transient states not\neasily obtained by standard MD simulations, in conformity with nuclear\nmagnetic resonance structural indications. This method offers an efficient\nand rational framework for comprehensive protein conformational sampling,\ncontributing significantly to our understanding of protein dynamics\nand function.