neuromodes.connectome.model_connectome

neuromodes.connectome.model_connectome(emodes: ArrayLike, evals: ArrayLike, r: float = 9.53, k: int = 108) → NDArray

Generate a vertex-wise structural connectivity matrix using the Geometric Eigenmode Model [1].

Parameters

emodesarray-like: The eigenmodes array of shape (n_verts, n_modes), where n_verts is the number of vertices and n_modes is the number of eigenmodes.
evalsarray-like: The eigenvalues array of shape (n_modes,).
rfloat, optional: Spatial scale parameter for the Green’s function, in millimeters. Default is 9.53.
kint, optional: Number of eigenmodes to use. Default is 108.

Returns

np.ndarray: The generated vertex-wise structural connectivity matrix.

Raises

ValueError: If emodes does not have shape (n_verts, n_modes) where n_verts ≥ n_modes.
ValueError: If evals does not have shape (n_modes,).
ValueError: If r is not a positive number.
ValueError: If k is not a positive integer in the range [1, n_modes].

Notes

If comparing this model to empirical connectomes, consider thresholding the returned matrix to match the density of the empirical data.

References