Two-photon calcium imaging day recordings (columns) of five neurons (rows) tracked with the open-source algorithm Track2p. The first day corresponds to postnatal day 8 and the last to P14. Cells shown in white with a blue center. Image credit: Majnik et al. (CC BY 4.0)
In the weeks and months after birth, the brain experiences rapid growth and restructuring through carefully timed developmental sequences. These processes rely heavily on coordinated neuronal activity for the brain to develop healthily.
Understanding these processes requires recording neural activity from the same cells across different stages of development. However, this has proven technically challenging because both the anatomy and cellular organization of the brain change dramatically during early life.
Two-photon calcium imaging, which uses fluorescent markers to visualize neuronal activity in living animals, has emerged as a powerful method for studying neural circuits. In adult animals, automated techniques can track large neuronal populations across days. However, the rapid brain growth and morphological changes during development make it difficult to track neurons during this stage. Recording from the same developing neurons would provide a more dynamic view of healthy brain maturation and reveal how it diverges in neurodevelopmental conditions.
Majnik et al. investigated whether it was possible to automatically and reliably track the same neurons across consecutive days during early postnatal development of mice. They developed Track2p, an open-source algorithm that can track developing neurons across days using a two-step procedure. It first corrects for brain growth using image registration and then matches neurons across days using this growth-corrected alignment.
Applying Track2p to calcium imaging data from the mouse barrel cortex during the second postnatal week, Majnik et al. found that the algorithm robustly tracked hundreds of neurons despite substantial brain growth. Benchmarking against manually tracked neurons confirmed high accuracy. Analysis of the tracked population’s activity properties revealed an increase in overall activity rates and a decrease in firing synchrony. Moreover, around postnatal day 11, neuronal activity patterns shifted from highly synchronized and spatially organized population events to more decorrelated, behavior-dependent firing.
Majnik et al. demonstrate that Track2p overcomes key technical barriers and reveal new principles of early postnatal brain maturation. The tool enables longitudinal analysis of neural circuits, allowing researchers to measure how individual neurons develop over time. Tracking cells across days will be particularly useful for understanding how genetic mutations or environmental factors influence developmental trajectories. Beyond development, Track2p can also examine long-term properties of adult circuits, such as learning or representational stability. Ultimately, applying Track2p to disease models may identify early circuit-level biomarkers of neurodevelopmental disorders.
