Synthesis of geometrically realistic and watertight neuronal ultrastructure manifolds forin silicomodeling

Abstract

Understanding the intracellular dynamics of brain cells entails performing three-dimensional molecular simulations incorporating ultrastructural models that can capture cellular membrane geometries at nanometer scales. While there is an abundance of neuronal morphologies available online, e.g. fromneuromorpho.org, converting those fairly abstract point-and-diameter representations into geometrically realistic and simulation-ready, i.e. watertight, manifolds is challenging. Many neuronal mesh reconstruction methods have been proposed, however, the resulting models are either biologically unplausible or non-watertight. We present an effective and unconditionally robust method capable of generating geometrically realistic and watertight surface manifolds of spiny cortical neurons from their morphological descriptions. The robustness of our technique is assessed with a mixed dataset of cortical neurons with a wide variety of morphological classes. The implementation is seamlessly extended and applied to synthetic astrocytic morphologies that are also plausibly biological in detail. Resulting meshes are ultimately used to create tetrahedral models that are plugged intoin silicoreaction-diffusion simulations for revealing cellular structure–function relationships.Availability and implementationOur method is implemented inNeuroMorphoVis, a neuroscience-specific open sourceBlenderadd-on, making it freely accessible for neuroscience researchers.Key pointsA plethora of neuronal morphologies is available in point-and-diameter formats, but there are no robust methods to convert these morphologies into realistic geometric models to conduct subcellular simulations based on synaptic data emerging from digitally reconstructed neuronal circuits.We present a scalable method capable of creating high fidelity watertight ultrastructural manifolds of complete neuronal models from their one-dimensional descriptions.Resulting manifold models include geometrically realistic somata and spine geometries, enabling accuratein silicoexperiments that can probe intricate structure-function relationships.Our method is extensible and can be seamlessly applied to astroglial morphologies and large networks of cerebral vasculature.