Jazmine Brash visited the Roslin Institute to develop DeepLabCut-based methods for assessing whether implanted biosensors alter gait in sheep. The same AI-based approach was also used to compare healthy sheep with an ovine model of Batten disease, a fatal inherited neurodegenerative disorder that affects both sheep and humans. The analysis found no evidence that the implanted biosensors altered normal gait. In contrast, the method detected clear gait differences between healthy and Batten disease-affected sheep. The project demonstrates the sensitivity of AI-based gait analysis for objectively detecting changes in gait. Such phenotyping tools can complement advances in genome engineering by enabling more precise characterisation of genetically engineered livestock and large-animal disease models. 1. Can you briefly describe your project at Roslin and how visiting the Roslin Institute supported your research?This project involved analysing sagittal video recordings from twenty 18-month-old sheep, 10 of which had a biosensor health monitor surgically implanted 12 months earlier. The video data were collected at Lincoln University and later transferred to the Roslin Institute, where I used DeepLabCut (DLC) software to perform AI-based gait analysis.Once the DLC software had been trained to recognise key anatomical points that we identified as important indicators of gait abnormality, it was used to automatically label and track movement throughout the video dataset, generating trajectories and large CSV files containing positional coordinate data. The coordinate data were then analysed in R software to compare gait patterns between the implanted and non-implanted sheep. No significant differences were detected between the two groups, indicating that the biosensor health monitors did not affect sheep gait.A second objective of the project was to determine whether gait differences could be identified between healthy control sheep and naturally occurring CLN5 and CLN6 sheep model of Batten disease. At 18 months of age (disease endpoint), a noticeable difference in gait was detected. 2. What unique resources, such as equipment, expertise, or bioresources did you access at Roslin, and how did these enhance your project?While at the Roslin Institute, I had access to a high-specification laptop equipped with the required DLC and R software needed for video gait analysis. These resources were essential for processing and analysing the large volumes of video data generated during the project, with individual analyses often taking several days to complete. Access to this equipment and software significantly improved the efficiency and reliability of the data analysis process. 3. How did working alongside Roslin’s research staff (technicians, postdocs, PIs) shape your experimental design, data analysis, or overall scientific thinking?This project would not have been possible without the expertise available at the Roslin Institute. DLC is a highly specialised software platform that requires coding knowledge and experience to implement effectively. Working alongside researchers who were already experienced with DLC and actively developing solutions to coding challenges was invaluable to the success of the project.Due to the technical nature of the work, the project became a close collaboration between Dr Sam Eaton, Dr Daniel Tolhurst, and myself. Their expertise helped shape the data analysis workflow, troubleshoot technical issues, and improve the overall approach to the project. Through these interactions, I gained a greater understanding of AI-based video analysis, coding applications in research, and the importance of interdisciplinary collaboration when tackling complex scientific questions. Gait analysis in sheep using DeepLabCut. A representative video frame shows the identification and tracking of anatomical landmarks, enabling quantitative analysis of posture and gait characteristics. 4. As an early-career researcher, what were the key professional and personal benefits, such as new skills, publications, or international exposure, that you gained from the experience?This trip and project were beneficial on multiple levels, contributing to both my personal and professional development. Through the project, I expanded my knowledge of available research tools while developing and refining essential technical and analytical skills. These experiences also created opportunities for future publications and collaborations with researchers working in overlapping fields and shared areas of interest.Beyond the direct research outcomes, the project resulted in the development of a video collection guide and step-by-step manuals for the required DLC and R software, allowing the methodology to be easily replicated in future gait analysis studies. The collaborative nature of the experience also strengthened professional connections and interdisciplinary learning opportunities, helping to build a broader research network and increasing future accessibility for collaborative work. 5. What advice would you give to researchers interested in applying to attend a Visiting Researcher Programme at Roslin?The Roslin Institute is a highly collaborative environment that makes it easy to interact with researchers and feel involved in the wider research community. Whether through informal discussions, impromptu meetings, or simply asking questions, there are many opportunities to learn from the people around you and engage with the work being carried out across the institute.To get the most out of the experience, it is important to be proactive, interactive, and willing to ask questions. Sharing your own experiences and expertise can often lead to interesting discussions and new perspectives, while also helping to build professional connections. In my experience, the people at Roslin were approachable, welcoming, and genuinely interested in both me personally and the research project I was involved in. This article was published on Tuesday 27 October 2026
