Computational methods and fast hybrid stochastic simulations for triangulation sensing and identifying principles of cell navigation in the brain

Brownian simulations can be used to generate statistics relevant for studying molecular interactions or trafficking. However, the concurrent simulation of many Brownian trajectories at can become computationally intractable. Replacing detailed Brownian simulations by a rate model was the basis of Gillespie's algorithm, but requires one to disregard spatial information. However, this information is crucial in molecular and cellular biology. Alternatively one can use a hybrid approach, generating Brownian paths only in a small region where the spatial organization is relevant and avoiding it in the remainder of the domain. Here we review the recent progress of hybrid methods and simulations in the context of cell sensing and guidance via external chemical gradients. Specifically, we highlight the reconstruction of the location of a point source in 2D and 3D from diffusion fluxes arriving at narrow windows located on the cell. We discuss cases in which these windows are located on the boundary of the 2D or 3D half-space, on a disk in free space, inside a 2D corridor, or a 3D ball. The hybrid method in question performs Brownian simulations only inside a region of interest. It uses the Neumann-Green's function for the mentioned geometries to generate exact mappings exit and entry points when the trajectory leaves the region, thus avoiding the explicit computation of Brownian paths in an infinite domain. Matched asymptotics is used to compute the probability fluxes to small windows and we review how such an approach can be used to reconstruct the location of a point source and estimating the uncertainty in the source reconstruction due to an additive perturbation present in the fluxes. We also review the influence of various window configurations on the source position recovery. Finally, we discuss potential applications in developmental cell biology and possible computational principles.