Simulation Data - In silico model of axonal pathfinding during spinal cord regeneration in zebrafish larvae

Overview This dataset contains the agent-based model (ABM) simulation outputs associated with the revised manuscript In silico model of axonal pathfinding during spinal cord regeneration in zebrafish larvae by Oskar Neumann, Mathar Kravikass, Nora John, Rahul Gopalan Ramachandran, Paul Steinmann, Vasily Zaburdaev, Daniel Wehner and Silvia Budday. The model investigates mechanically guided axonal growth through a deformable extracellular matrix (ECM). Local ECM stiffness influences axonal growth through phenomenological coupling functions that modulate the effective link range and growth cone angle. The simulations investigate the influence of different stiffness distributions, mathematical coupling functions and mechanosensing parameter combinations on the resulting axonal growth patterns. This repository contains the underlying simulation data. The simulation outputs can be used for visualization, trajectory analysis and further post-processing. Simulation output format Each simulation is stored as a series of plain-text snapshot files (snapXXX.txt), representing recorded simulation states. The numerical suffix identifies the snapshot, with the highest-numbered file corresponding to the final recorded simulation state. Each snapshot contains information about individual particles, including their positions, types, radii and identifiers. The model distinguishes ECM particles, neurons, neurite particles and neurite leading beads. Repository structure The repository contains six ZIP archives organized according to the corresponding figures in the revised manuscript. Figure 7.zip: Stiffness-pattern test cases Simulations investigating axonal growth under five different ECM stiffness distributions: Horizontal bands with decreasing stiffness from top to bottom. Radial stiffness distribution with a soft center. Radial stiffness distribution with a stiff center. Cross-shaped distribution with intersecting stiff diagonal bands. Cascading stiffness-band distribution. The simulations investigate four mechanosensing conditions: no explicit mechanosensing, stiffness-modulated link range with straight growth, stiffness-modulated directionality, and both mechanisms combined. Figure 8.zip: Mechanosensing parameter sensitivity A systematic parameter sweep investigating the influence of the link-range sensitivity parameter β_link and the directional sensitivity parameter n on axonal growth within the cascading ECM stiffness distribution. The simulations cover 25 parameter combinations, with: β_link = [0, 2.5, 7.5, 12.5, 25] n = [0, 0.5, 2.5, 5, 10] Subdirectories are named beta_XX_n_YY, where numerical values are multiplied by ten. For example, beta_75_n_25 corresponds to β_link = 7.5 and n = 2.5. Figure S5.zip: Original mechanosensing coupling functions Sixteen simulations investigating different mathematical forms of the two mechanosensing coupling functions on the cascading stiffness distribution. Both the link-range and directional couplings are systematically varied among four functions: logistic, power law, linear and smoothstep. Subdirectories follow the naming convention beta_<function>_n_<function>, with the abbreviations log, pow, lin and step. The nominal sensitivity parameters are β_link = 7.5 and n = 5.0. Figure S6.zip: Inverted mechanosensing coupling functions Sixteen simulations investigating the inverted versions of the four original coupling functions. The investigated functions comprise inverted logistic, inverted power law, inverted linear and inverted smoothstep mappings. All combinations of the four inverted link-range and four inverted directional coupling functions are included. Subdirectories follow the same naming convention as Figure S5, with the additional prefix inv_ identifying inverted functions. Figure S7.zip: Non-monotonic mechanosensing coupling functions Nine simulations investigating combinations of reference and non-monotonic coupling functions on the cascading stiffness distribution. The link-range coupling functions comprise the original logistic, truncated parabolic and Gaussian mappings. The directional coupling functions comprise the original power law, truncated parabolic and Gaussian mappings. Subdirectories use the abbreviations log, pow, par and gauss to identify the corresponding mathematical functions. Figures 9 and S12.zip: Progressively evolving wound-edge simulations Simulations of axonal regrowth through a prescribed, progressively evolving ECM stiffness distribution representing a stiff wound edge. The simulations investigate two initial growth cone angles, α = 30° and α = 75°, and combinations of the following mechanosensing parameters: β_link = [2.5, 7.5, 12.5] n = [0.5, 2.5, 5.0] Five independent stochastic simulation realizations were performed for each parameter combination. These simulations provide the underlying data for the representative axonal growth patterns in Figure 9 and the selected simulation realizations presented alongside experimental images in Supplementary Figure S12. The same simulation data can also support additional analyses, including mean squared displacement, tortuosity and spatial signal-density profiles, as presented elsewhere in the revised manuscript. Scope of the dataset This repository provides simulation outputs only. Experimental microscopy images, generated manuscript figures and post-processing scripts are not included. The simulation snapshots can serve as input for independent visualization and post-processing, including analyses of axonal trajectories, spatial growth patterns and stochastic variability. Reproducing comparisons with experimental fluorescence profiles requires the corresponding experimental data and analysis procedures described in the manuscript. Version information This revised dataset reorganizes the original simulation archives to match the revised manuscript and extends the original repository with the inverted and non-monotonic coupling-function simulations.