Summary: A new 3D platform that simulates zebrafish behavior opens new avenues for research while reducing reliance on live animals.
Source: NYU Tandon School of Engineering.
Each year, roughly 20 million animals are used in scientific studies. Zebrafish are increasingly favored over rodents and primates because of their genetic similarities to humans, rapid development, and experimental versatility. Growing ethical concerns from policymakers, the public, and scientific bodies have driven a search for alternatives that minimize animal use. Computer-based, data-driven models promise to reduce the number of live animals required for many types of experiments without sacrificing scientific rigor.
A research team led by Maurizio Porfiri, professor of mechanical and aerospace engineering at the NYU Tandon School of Engineering, has created the first data-driven modeling framework capable of simulating zebrafish swimming behavior in three dimensions. Built from experimental observations, this 3D in-silico model is robust enough to support certain preclinical and neurobehavioral studies that historically depended on living fish.
The research, published in Scientific Reports, is titled “In-silico Experiments of Zebrafish Behavior: Modeling Swimming in Three Dimensions.” The paper’s authors are Maurizio Porfiri, NYU Tandon doctoral candidate Violet Mwaffo, and Sachit Butail, an assistant professor at Northern Illinois University.
Extending earlier two-dimensional models developed in 2015, the team incorporated advances inspired by techniques from financial engineering—an area where coauthor Violet Mwaffo has training—to create a richer behavioral representation. The new 3D model includes burst-and-coast swimming, speed modulation, and interactions with tank walls. These features enable realistic virtual experiments that can reproduce common behavioral measures researchers use to study locomotion and neurobehavioral responses.
The model was calibrated using a novel three-dimensional tracking dataset collected by Porfiri’s group. That tracking framework was developed by Sachit Butail during his postdoctoral work at NYU Tandon and provided the high-quality motion data needed to tune the model’s parameters.
To validate the model, the researchers compared its behavior to published experimental measurements of zebrafish swimming speeds across tanks of different sizes. Both literature data and the model revealed a clear correlation: fish swim faster in larger tanks. Detecting such relationships experimentally would normally require thousands of animals and months of work; the computational approach can establish and test these trends in minutes of simulation time.
Porfiri emphasizes that the model is not intended to replace all animal testing but to serve as an effective tool during preclinical stages. “We’re proposing to use this zebrafish model in early-stage research to reduce the number of animals needed for hypothesis screening and experimental design,” he said. By identifying promising conditions computationally, laboratories can design fewer, more focused animal experiments when live testing is unavoidable.
Although the current model reproduces many key ethological observables—such as spatial distribution, speed profiles, and wall-following behaviors—work remains to capture the full range of zebrafish behaviors. Future developments will extend the framework to simulate social interactions, group dynamics, and responses to live or engineered stimuli, enabling richer virtual studies of schooling, attraction, and sensory-driven behavior.
Funding: This research was supported by the National Science Foundation.
Source: Kathleen Hamilton, NYU Tandon School of Engineering.
Image Source: Image credited to NYU Tandon School of Engineering.
Original Research: Research article “In-silico Experiments of Zebrafish Behaviour: Modeling Swimming in Three Dimensions” by Violet Mwaffo, Sachit Butail, and Maurizio Porfiri, published in Scientific Reports on January 10, 2017 (doi:10.1038/srep39877).
Suggested citation: NYU Tandon School of Engineering. “Swimming for Science: 3D Simulations of Zebrafish Behavior May Reduce the Need for Live Animals in Some Research.” NeuroscienceNews, January 12, 2017.
Abstract
In-silico experiments of zebrafish behaviour: modeling swimming in three dimensions
Zebrafish are rapidly becoming a preferred model in biomedical research for studying normal and disrupted biological processes, especially when coupled with genetic and pharmacological manipulation. Like mammalian models, zebrafish experimentation raises ethical concerns that motivate the development of alternative approaches to reduce the number of live animals, refine experimental protocols, and, where possible, replace animal use. Building on successful computational methods in other life sciences, we developed a data-driven framework to simulate zebrafish swimming in three dimensions. The model incorporates burst-and-coast swimming, speed modulation, and wall interaction, providing a platform for in-silico experiments. Computational studies with this model reproduce common behavioral metrics such as swimming speed and spatial preferences, and predict how tank dimensions influence behavior. By enabling efficient virtual testing, this framework can help reduce animal use and suffering while supporting rigorous scientific discovery.
“In-silico experiments of zebrafish behaviour: modeling swimming in three dimensions” by Violet Mwaffo, Sachit Butail, and Maurizio Porfiri, Scientific Reports, January 10, 2017. doi:10.1038/srep39877