A Fine Motor Task to Study Joint Kinematics in a Preclinical Model of Neurodegenerative Disease
June 13th, 2025
The current study describes a fine motor behavior test for examining motor deficits in rodent models, including the TgF344-AD rat, using machine learning.
Our research focuses on fine motor skills in models of neurodegeneration. Specifically, we aim to use our new task as a novel biomarker for motor dysfunction. Machine learning algorithms have advanced systems neuroscience to provide an unbiased and automatic way of analyzing clustered animal movement into behavior that typically cannot be deciphered by the human eye.
By combining body part tracking and computer clustering methods, we are able to reliably detect, classify, and predict transitions between animal movements in our novel fine motor task. To begin the 10-day shaping phase, first turn on the infrared light, the camera, and the computer. Transfer the cages containing the rats into the testing room and allow them to acclimate for five minutes.
Using the training bowl, provide one to five pellets along the extension with the extension placed fully inside the apparatus on the first day. Conduct 10-minute training sessions with each animal. On subsequent days, place the pellets at the end of the bowl closest to the rodent and gradually move them farther away to encourage reaching behavior.
Once rodents take pellets from the edge of the extension, even if only with their mouths, introduce pellet presentation with tweezers to promote reaching behavior. Reward the rodent immediately upon touching the tweezers. Then, hold the tweezers slightly farther to prompt the rodent to reach further.
For the nine-day testing phase, place the rats in the room to habituate for one hour. Apply dental wax to the platform to fix it in place, then position the first bowl, ensuring the front is flush and centered with the apparatus opening. Name the video file and Spin View with the rat identification number, date of recording, and time of recording.
Transfer the rat into the apparatus. Close the lid and position a white cardstock on top to deflect light. Double-check the camera angle to ensure the appropriate scene is captured and adjust the bowl if it is not centered.
Pour the specified number of pellets into the bowl and press Start Recording. Stay in the room to monitor the rat, take behavioral notes, and replenish pellets as needed. For the Plain bowl, ensure pellets are visible and accessible over the bowl edge.
For the Less bowl, verify that pellets are not stacked and remain loosely scattered. For the Plinko bowl, check that pellets are adequately distributed between obstructions. Return the rat to its home cage and provide the allotted food.
Wipe the chamber with a disinfectant between animals. At the end of the day, spray the entire chamber with ethanol for disinfection. During baseline testing, rats showed increased total reaches as the bowl design became more complex.
Under Harmaline treatment, total reaches decreased across all bowl types compared to baseline, indicating impaired motor function. During baseline, total reaches increased with bowl difficulty, showing behavioral persistence despite task complexity. Failures increased with the Plinko bowl.
Harmaline disrupted typical reach performance. Prior to Harmaline treatment, failures increased with task difficulty under baseline, but remained consistent under Harmaline. Harmaline significantly reduced the learning rate of the task and reaching rate across all bowl types.
Analysis of reaches using machine learning revealed unique poses and task strategy between baseline and Harmaline conditions and within each bowl type. Distinct reaching syllables were used by rats within each bowl type and during different bowl conditions. TgF344-AD transgenic rats had reduced total reaches compared to matched wild-type rats.
Wild-type rats showed increased total reaches with bowl complexity. While performance type was variable within each group, the TgF344-AD rats had significantly reduced success performance. As bowls became more challenging, the number of failures increased in both genotypes.
Behavior can create a lot of noise. In particular, with our task where pellets can be knocked off the platform. We solve this problem with keypoint-MoSeq.
Our protocol improves upon past research behavior paradigms that do not allow for self-perturbing features of movement.
View the full transcript and gain access to thousands of scientific videos
Overview
This study introduces a fine motor behavior test designed to assess motor deficits in rodent models, particularly the TgF344-AD rat, utilizing machine learning techniques.
Key Study Components
Area of Science
- Neuroscience
- Behavioral Analysis
- Machine Learning
Background
- Focus on fine motor skills in neurodegeneration models.
- Development of a novel biomarker for motor dysfunction.
- Advancements in machine learning for analyzing animal movement.
- Importance of unbiased analysis in behavioral studies.
Purpose of Study
- To create a reliable test for assessing motor deficits.
- To utilize machine learning for behavior classification.
- To enhance understanding of motor dysfunction in rodent models.
Methods Used
- Combination of body part tracking and computer clustering methods.
- Detection and classification of animal movements.
- Implementation of a 10-day shaping phase for acclimatization.
- Use of infrared light and camera for observation.
Main Results
- Reliable detection of transitions between animal movements.
- Successful classification of fine motor tasks in rodents.
- Demonstrated potential for machine learning in behavioral analysis.
- Establishment of a new approach for studying motor dysfunction.
Conclusions
- The fine motor behavior test is effective for assessing motor deficits.
- Machine learning provides a powerful tool for behavioral analysis.
- This approach may lead to better understanding of neurodegenerative conditions.
Frequently Asked Questions
Executive Industry Relevance
Quantitative assessment of fine motor dysfunction in preclinical neurodegenerative models addresses a critical gap in early-stage target validation and mechanistic de-risking for Alzheimer's and related disorders. The integration of machine learning-based kinematic analysis enables objective, reproducible measurement of motor phenotypes, supporting predictive confidence in translational research. This approach enhances portfolio decision-making by providing sensitive, scalable endpoints for disease progression and therapeutic intervention studies.
Strategic Applications in Biopharma R&D
Early Discovery & Target Validation
- Enables objective interrogation of motor phenotypes linked to neurodegenerative pathology.
- Supports functional target validation by quantifying disease-relevant behavioral deficits.
- Facilitates mechanistic de-risking through automated detection of subtle motor impairments.
- Provides reproducible endpoints for early-stage portfolio triage.
Screening & Assay Development
- Establishes validated behavioral assays for downstream compound screening in rodent models.
- Delivers standardized, quantitative outputs for cross-study comparability.
- Enables reproducible assessment of intervention effects on motor function.
- Supports scalable assay deployment using open-source machine learning tools.
Translational & Preclinical Research
- Aligns preclinical endpoints with human-relevant motor and cognitive biomarkers.
- Enables longitudinal tracking of disease progression and therapeutic response.
- Supports risk-adjusted advancement decisions based on sensitive behavioral readouts.
- Facilitates continuity from discovery through preclinical validation in neurodegeneration pipelines.
Pipeline & Workflow Integration
This kinematic motor task integrates into the discovery-to-preclinical continuum, providing a bridge from mechanistic studies to translational biomarker development in neurodegenerative disease research.
- Discovery Biology: Supports hypothesis testing and pathway clarification by quantifying motor deficits in disease models.
- Screening: Delivers assay-ready, reproducible behavioral endpoints for compound evaluation.
- Analytics: Provides quantitative, machine learning-derived metrics for robust statistical comparison across conditions.
- Translational Research: Aligns preclinical motor assessments with emerging human biomarkers for early disease detection.
- Enterprise Reuse: Offers a scalable, open-source platform adaptable across neurodegenerative models and research sites.
Operational & Enterprise Impact
- Scientific Value: Increases predictive confidence and reduces mechanistic ambiguity in neurodegeneration research.
- Operational Value: Standardizes behavioral phenotyping and enhances reproducibility across studies.
- Strategic Value: Informs go/no-go decisions and optimizes resource allocation in early-stage pipelines.
- Portfolio Impact: Enables risk-adjusted prioritization of therapeutic candidates targeting motor dysfunction.
Implementation Considerations
- Requires expertise in behavioral neuroscience and machine learning analytics.
- Needs video capture infrastructure and computational resources for data processing.
- Demands cross-team standardization of scoring and analysis protocols.
- Adaptable to various rodent models of neurodegeneration with protocol optimization.
- Potential limitations include behavioral variability and environmental sensitivity, mitigated by automated analysis.
