Data-driven stochastic modelling of zebrafish locomotion

Area of Science:

• * Computational Biology
• * Animal Behavior
• * Biophysics

Background:

• * Understanding fish locomotion is crucial for ecological and behavioral studies.
• * Previous models often lack detailed, data-driven representations of individual fish movement dynamics.
• * Zebrafish (Danio rerio) are a key model organism for studying collective and individual behaviors.

Purpose of the Study:

• * To develop a data-driven mathematical framework for modeling zebrafish locomotion.
• * To incorporate dynamic speed regulation and responses to external constraints into fish movement models.
• * To provide a quantitative tool for analyzing individual fish behavior in controlled environments.

Main Methods:

• * Utilized automated visual tracking to capture individual zebrafish movement data.
• * Developed a model using stochastic differential equations to represent fish as self-propelled particles.
• * Calibrated model parameters using salient metrics derived from experimental data, including speed and angular speed.

Main Results:

• * Successfully reproduced key characteristics of individual zebrafish locomotion.
• * Integrated experimentally-derived processes for dynamic speed regulation into the model.
• * Accounted for fish responses to external constraints within the mathematical framework.

Conclusions:

• * The developed data-driven framework offers a robust method for simulating fish movement.
• * This approach facilitates quantitative investigation of individual behavior under various experimental conditions.
• * The model provides insights into the fundamental dynamics governing zebrafish locomotion.