RahmaniPeruaniRomanczuk_PLoS_Comp_Biol_2020.zip

Data from an individual-based model generated by numerical integration of the underlying stochastic equations of motion. Individual agents moving with constant speed, interact with k nearest objects in their surrounding. Elements of the k nearest object set can be either other agents (social cues) or distraction sites (local environmental cues), which for simplicity are referred to as obstacles. If an agent has an obstacle within its k nearest objects and it is within its repulsion distance, the agent turns away. Otherwise, the agent coordinates its motion with other agents through an alignment interaction. A small fraction of agents has additional private information about a global direction of motion and biases its motion accordingly. This data set contains all data required to generate figures of the linked manuscript. For details on the mathematical model, see corresponding Materials and Methods section in the manuscript.<br><br>Data was obtained from numerical simulations of the individual-based model by solving the stochastic differential equations of motion using custom made simulation code written in C++. Data contains depending on the file/figure:<br>individual coordinates of agents in 2D (x,y,vx,vy) or coordinates of (stationary) obstacles (x,y);<br>average quantities for individual runs for different parameter values (migration accuracy, obstacle avoidance, etc).<br>2D histogram data for the 2D probability density of in and out degrees of the interaction network.<br>See linked manuscript for the definition of the average quantities.<br><br>See included README.txt for additional information of the data contained in individual files. In addition two IPython notebooks are provided which contains additional processing steps and information for generation of corresponding plots (as in the linked manuscript).