neuromaps.nulls.moran

neuromaps.nulls.moran(data, atlas='fsaverage', density='10k', parcellation=None, n_perm=1000, seed=None, distmat=None, n_proc=1, **kwargs)[source]

Generates null maps for data using method from [SN9]

Method uses a spatial decomposition of a distance-based weight matrix to estimate eigenvectors that are used to generate surrogate maps by imposing a similar spatial structure on randomized data. For a MATLAB implementation refer to [SN10] and [SN11]

Parameters
  • data ((N,) array_like) – Input data from which to generate null maps

  • atlas ({'fsLR', 'fsaverage', 'civet'}, optional) – Name of surface atlas on which data are defined. Default: ‘fsaverage’

  • density (str, optional) – Density of surface mesh on which data are defined. Must be compatible with specified atlas. Default: ‘10k’

  • parcellation (tuple-of-str or os.PathLike, optional) – Filepaths to parcellation images ([left, right] hemisphere) mapping data to surface mesh specified by atlas and density. Should only be supplied if data represents a parcellated null map. Default: None

  • n_perm (int, optional) – Number of null maps or permutations to generate. Default: 1000

  • seed ({int, np.random.RandomState instance, None}, optional) – Seed for random number generation. Default: None

  • n_proc (int, optional) – Number of processors to use for parallelizing computations. If negative will use max available processors plus 1 minus the specified number. Default: 1 (no parallelization)

  • distmat (tuple-of-str or os.PathLike, optional) – Filepaths to pre-computed (left, right) surface distance matrices. Providing this will cause atlas, density, and parcellation to be ignored. Default: None

  • kwargs (key-value pairs) – Other keyword arguments passed directly to the underlying null method generator

Returns

nulls – Generated null distribution, where each column represents a unique null map

Return type

np.ndarray

References

SN9

Wagner, H. H., & Dray, S. (2015). Generating spatially constrained null models for irregularly spaced data using M oran spectral randomization methods. Methods in Ecology and Evolution, 6(10), 1169-1178.

SN10

de Wael, R. V., Benkarim, O., Paquola, C., Lariviere, S., Royer, J., Tavakol, S., … & Bernhardt, B. C. (2020). BrainSpace: a toolbox for the analysis of macroscale gradients in neuroimaging and connectomics datasets. Communications Biology, 3(1), 1-10.

SN11

https://github.com/MICA-MNI/BrainSpace/