Neural Signals and Connectomes

This lecture explores neural signals and connectomes, focusing on graph theory to model brain networks. It covers the structural and functional connectomes, graph measures, models, and Laplacian. The instructor discusses modern connectomics, fruit fly and mouse connectomes, and the relevance of functional connectomes in neurodevelopment. The lecture delves into graph partitioning, basic measures, clustering coefficients, and path lengths. It also examines models like Erdös-Rényi, Barabási-Albert, and Watts-Strogatz, and explains graph partitioning through convex relaxation and Fiedler vector. The session concludes with multi-voxel pattern analysis in fMRI trials, linear classification, and decoding models.