Functional Connectivity
GLM Toolbox
A toolbox for fitting generalized linear models of multiple spike-train data. Includes methods for fitting coupled and uncoupled models, with and without external covariates, with L1 or L2 regularization and a variety of parameterizations (see Pillow et al. (2008) and Truccolo et al. (2005)).
Code: coming soon
Reference: PDF Inferring functional connections between neurons (2008)
Stevenson IH, Rebesco JM, Miller LE, and Körding KP.
Current Opinion in Neurobiology. in press.
Hierarchical GLM with sparseness and smoothness priors
A toolbox for fitting the functional connectivity model described in Stevenson et al. (2008). It uses coordinate ascent and the RPROP (resilient back-propagation) algorithm to fit functional connections and connection weight parameters from a set of spike trains. This toolbox also contains wrappers for clustering the connectivity matrix. Code and references for the IRM (Infinite Relational Model) algorithm can be found at Charles Kemp's website.
Code: coming soon
Reference: PDF Bayesian inference of functional connectivity and network structure from spikes (2008)
Stevenson IH, Rebesco JM, Hatsopoulos NG, Haga Z, Miller LE, Körding KP.
IEEE Transactions on Neural Systems and Rehabilitation Engineering, Special Issue on Brain Connectivity.
