使用三室开放式自由游动实验检测斑马鱼（Danio rerio）社会偏好行为的低成本替代方法
2026年8月7日
多室社交偏好实验已从啮齿类动物改良应用于斑马鱼，有助于在模式生物中研究社交行为。本方案验证了一种低成本的三室开放式水箱实验方法，利用网络摄像头和开源软件测量斑马鱼的社交偏好，使其更适用于本科生科研与教学。
本方案验证了一种低成本、开源的三室开放式水箱自由游泳实验，用于研究斑马鱼的社会偏好行为。该方案可适用于多种物种，并允许本科生研究人员在资源有限的条件下开展社会偏好研究。首先，准备一个包含中央实验区和两侧刺激区的开放式水箱装置。
安装背光系统。使用不透明隔板和偏振滤光片以优化视觉隔离和视频对比度。使用相机支撑架将相机安装并校准在腔室前方。
拉上遮光窗帘，并在开始实验前保持稳定的环境条件。下载 Open Broadcaster 软件并将其安装在运行 Windows 操作系统的本地计算机上。启动软件后，运行自动配置向导，以根据特定的录制需求、硬件资源和网络条件优化设置。
在配置摄像头之前，验证基本设置。将基础画布分辨率设置为与显示分辨率相匹配，并设置适当的输出缩放分辨率。然后，通过在场景框中右键单击并选择“添加场景”来创建一个场景。
为场景指定一个名称，并使用该名称定义视频流布局。添加新的源，选择“视频采集设备”以添加网络摄像头。为每个摄像头分配一个唯一的名称，例如 Camera Center，并对其他摄像头重复此过程。
在预览窗口中调整源的位置和大小。确保摄像头列于源列表中。按住 Control 键并按下 D 键，将摄像画面移至画布中央。按住 Control 键并按下 F 键以适应屏幕，或按住 Control 键并按下 S 键拉伸画面。按住 Alt 键的同时拖动边界框，以裁剪源画面。
接下来，双击选中的摄像头，然后点击“配置视频”。在属性菜单中调整视频的亮度和对比度。从养殖箱中向三个腔室注满水，并在引入鱼类前调节水温。
在整个适应期间，确保各 chamber 之间的不透明屏障已放置到位。将一条成年斑马鱼（Danio rerio）转移至中央 chamber，并使其适应至少一小时。将人工社交刺激物放置在左侧相邻的刺激 chamber 中。
然后，将由两条大小、年龄和熟悉程度相似的雄鱼和雌鱼组成的小群体（共四条鱼）转移至右侧刺激室。在适应期间，为所有含有活体动物的室提供设定在26摄氏度的浸入式加热器、连接水族箱气泵的气石以及水箱盖，以维持稳定的饲养条件。在开始实验前，移除加热器和气石。
启动中央水箱的单个摄像头。运行软件，通过点击并拖动视频角落将中央摄像头画面调整至全屏显示。然后，在控制选项中选择“开始录制”。
立即移除不透明屏障。10分钟后，通过在控制选项中选择“停止录制”来结束会话。通过点击左上角菜单中的“文件”，并在下拉菜单中选择“显示录制内容”，即可访问已录制的会话。
将四到五个视频载入运动分析软件应用程序（例如 DeepLabCut），并从每个视频中提取 20 至 30 帧。定义至少三个主要追踪点，例如头部、尾部和躯干。训练网络以在各帧中追踪这些点。
使用基于 ResNet-101 的神经网络训练模型。确定完成训练所需的迭代次数。随后，将完全训练好的 DeepLabCut 模型应用于其余实验视频，并按照用户指南提取每次试验的位置数据。
然后，使用 Python 等编程软件定义并分析在三个等宽垂直区域中的运动，以量化偏好。通过识别区域内至少两个身体点来测量每个区域中的停留时间，以减少追踪噪声。通过将区域内的总帧数除以总帧数，计算每个区域中的总时间。
使用相同的编码软件来定义和测量静止不动的时间，例如僵直状态。将运动阈值设定为每帧2到5个像素，且连续帧数至少为15帧。最后，汇总并导出测量结果，用于统计分析和可视化。
受试者在靠近活鱼群的右侧区域所花费的会话时间显著多于中间区域或靠近人工移动刺激物的左侧区域。在中间区域与左侧区域之间所花费时间的差异无统计学意义。受试者在中间区域保持静止的时间显著少于在靠近活鱼群的右侧区域或靠近移动刺激物的左侧区域。
本方案可供研究人员探究斑马鱼对两种不同社会性刺激的视觉互动倾向，并确定其对哪种刺激具有偏好。研究人员可采用不同物种或刺激组合方式，也可探究其他行为反应，例如快速冲刺或不规则游动。该系统为研究社会功能的转化模型提供了平台，还可用于探索药物学和环境因素的干预效应。
查看完整文字稿并访问数千部科学视频
Overview
This article presents a protocol for assessing social behavior in zebrafish using the 3-chamber open-tank free-swim task (OTFST). The method utilizes a cost-effective setup with USB webcams and common materials, combined with DeepLabCut for markerless pose estimation and behavioral tracking. The approach is validated through pilot sessions and is designed to be accessible for both research and teaching laboratories.
Key Study Components
Area of Science
- Neurobehavioral research
- Behavioral neuroscience
- Zebrafish behavioral analysis
Background
- Social preference tasks are widely used to study social behavior in animal models.
- The 3-chamber paradigm was originally developed for rodents and later adapted for zebrafish.
- Accurate behavioral tracking is essential for quantifying social interactions.
- Cost and accessibility can be barriers for widespread adoption of behavioral assays.
Purpose of Study
- To provide a low-cost, scalable protocol for assessing social preference in zebrafish.
- To demonstrate the use of DeepLabCut for accurate, markerless tracking of animal movement.
- To validate the system's ability to capture robust social behavior.
Methods Used
- 3-chamber open-tank free-swim task (OTFST) with a central tank and two stimulus tanks.
- Video recording using USB webcams positioned above the setup.
- Use of polarizing film and backlighting to enhance video quality.
- Application of DeepLabCut for markerless pose estimation and extraction of positional time-series data.
Main Results
- The protocol successfully captures social preference behavior in zebrafish.
- DeepLabCut provides accurate tracking data for behavioral analysis.
- The setup is validated through pilot sessions demonstrating robust behavioral metrics.
- The approach is budget-friendly and accessible for a wide range of laboratories.
Conclusions
- The 3-chamber OTFST protocol is effective for studying social interaction in zebrafish.
- The method is accessible, scalable, and suitable for both research and teaching environments.
- Open-source tools like DeepLabCut enhance the accuracy and reproducibility of behavioral studies.
Frequently Asked Questions
Executive Industry Relevance
Quantitative assessment of social behavior in zebrafish is increasingly relevant for neurobehavioral target validation and early-stage CNS drug discovery. The 3-chamber open-tank free-swim task (OTFST) with markerless video tracking enables scalable, reproducible behavioral phenotyping, supporting predictive confidence in translational models. This accessible, low-cost platform facilitates robust pipeline integration for teams prioritizing mechanistic de-risking and cross-study comparability.
Strategic Applications in Biopharma R&D
Early Discovery & Target Validation
- Enables interrogation of social behavior phenotypes relevant to neuropsychiatric target validation.
- Supports mechanistic de-risking by quantifying behavioral endpoints in a disease-relevant system.
- Facilitates portfolio triage by providing reproducible, quantitative behavioral data.
Screening & Assay Development
- Prepares validated zebrafish behavioral systems for downstream compound screening workflows.
- Standardizes assay conditions and outputs using open-source tracking and quantitative metrics.
- Enables scalable, reproducible behavioral assays for reliable compound evaluation.
Translational & Preclinical Research
- Aligns behavioral endpoints with translational biomarker strategies in CNS research.
- Supports continuity from early discovery through preclinical validation by enabling cross-study comparability.
- Provides predictive de-risking value for neurobehavioral phenotypes relevant to human disease.
Pipeline & Workflow Integration
The OTFST with DeepLabCut tracking fits within the early discovery to lead identification continuum, supporting hypothesis testing and quantitative behavioral analysis in zebrafish models.
- Discovery Biology: Enables robust hypothesis testing and pathway clarification for social behavior targets.
- Screening: Provides assay readiness and reproducibility through standardized video tracking and analysis.
- Analytics: Delivers quantitative positional and preference metrics for cross-condition comparison.
- Translational Research: Supports alignment of zebrafish behavioral endpoints with preclinical biomarker strategies.
- Enterprise Reuse: Offers a scalable, open-source platform adaptable across neurobehavioral research portfolios.
Operational & Enterprise Impact
- Scientific Value: Increases predictive confidence and reduces mechanistic ambiguity in neurobehavioral target validation.
- Operational Value: Standardizes behavioral assays with reproducible, scalable, and cost-effective infrastructure.
- Strategic Value: Improves go/no-go decision-making and capital efficiency by enabling robust behavioral phenotyping.
- Portfolio Impact: Supports risk-adjusted prioritization and advancement of CNS discovery programs.
Implementation Considerations
- Requires expertise in behavioral neuroscience and quantitative video analysis.
- Needs basic imaging infrastructure (USB webcams, lighting, and open-source software).
- Demands cross-team standardization of assay setup and data analysis protocols.
- Adaptable to other small aquatic models with protocol modifications.
- Practical limitations include video quality and environmental control for reproducibility.
Biopharma FAQs
Chapters in this video
0:00
Introduction
0:23
Testing Environment Setup
1:02
OBS Studio Configuration for Video Acquisition
2:42
Acclimation of Adult Zebrafish in the 3-Chamber Open-Tank
3:46
Experimental Social Preference Testing
4:30
Motion Tracking and Behavioral Analysis of Zebrafish Social Preference
6:13
Results
6:49
Conclusion
