Our lab mainly focuses on the neural circuits controlling sleep. Using optogenetic manipulation, optrode recording, and cell-type-specific calcium imaging, we identify neuronal types that play critical roles in the generation of rapid-eye-movement (REM) sleep and non-REM sleep.
This project will be focused on the lab mice behavioral classification (sleep, wakefulness (including locomotion, turning, feeding, rearing, grooming etc.)) with a un-supervised method. First, the behavioral videos of mice will be analysed with DeepLabCut to generate the mouse position coordinates in each frame (well established already). Then the data will be pre-processed for classification. The behavioral features will be extracted from the dataset, and then classified into different clusters. After that the machine learning or deep learning methods will be used for analysis all dataset to classify behaviors across all sleep and wake stages.
