Developing Multi-scale Data Analysis Tools and Modeling of Large-scale Neural Networks Mathematical Tools for Data-driven Neural Circuit Modeling

Recent advances in recording technologies are making it feasible to simultaneously record the activity of hundreds and thousands of neurons from behaving animals. However, methods to analyze the activity data of a large neural population are still to be developed. In this project, we develop mathematical technologies to uncover the spatiotemporal characteristics of neural population activity to read neural code and to model the circuit mechanisms underlying various brain functions.