Connectome analysis

This package provides connectome analysis routines.

class connectome.analysis.connectivity.ConnectivityEstimator(**kwargs)

Inputs: network, measures

Estimate the connectivity of a network.

Output Keys: p_ee, p_ii, p_ie, p_ei

class connectome.analysis.inoutdegreecorrelation.InOutDegreeCorrelation(**kwargs)

Inputs: network, measures

Estimates the correlation of in and out degrees of the excitatory subpopulation.

Output keys: in_out_degree_correlation_exc

class connectome.analysis.relativecycleanalysis.RelativeCycleAnalysis(**kwargs)

Inputs: length, network, measures

Compares relative to ER the expected number of cycles which is approximated here by \((p*n)^{length}\), where p is the connectivity, n the number of excitatory nodes and length the cycle length.

Note that this approximation ignores the non-independent nature of the Bernoulli product on the diagonal of the matrix power. This approximation fails for very small and very sparse networks but is precise for large and dense networks.

Configuration parameter: length, integer

Output keys: relative_cycles_<length>

class connectome.analysis.reciprocity.ReciprocityEstimator(**kwargs)

Inputs: network, measures

Estimate the network’s reciprocities; here “reciprocity_ei” means:

Given a connection from E -> I, what is the probability of the reciprocated connection to also exist?

Output keys: reciprocity_ee, reciprocity_ii, reciprocity_ei, reciprocity_ie

class connectome.analysis.relativereciprocity.RelativeReciprocityEstimator(**kwargs)

Inputs: network, measures

Estimate the reciprocity relative to an ER network of the same connectivity. See also ReciprocityEstimator.

Output keys: relative_reciprocity_ee, relative_reciprocity_ei, relative_reciprocity_ie, relative_reciprocity_ii