liana.mt.genericMistyData

liana.mt.genericMistyData(intra, intra_use_raw=False, intra_layer=None, extra=None, extra_use_raw=False, extra_layer=None, nz_threshold=0.1, add_para=True, spatial_key='spatial', set_diag=False, kernel='misty_rbf', bandwidth=100, zoi=0, cutoff=0.1, add_juxta=True, n_neighs=6, max_neighs=18, verbose=False)

Construct a MistyData object from an AnnData object with views as presented in the manuscript.

Parameters:

intra AnnData

AnnData object with the intraview

intra_use_raw bool (default: False)

Whether to use the raw data of the intraview.

intra_layer str | None (default: None)

The layer of the intraview to use.

extra AnnData | None (default: None)

AnnData object with the extraview(s). If None, the extraview is set to be the same as the intraview.

extra_use_raw bool (default: False)

Whether to use the raw data of the extraview.

extra_layer str | None (default: None)

The layer of the extraview(s) to use.

nz_threshold float (default: 0.1)

The threshold for the number of non-zero entries in each view.

add_para bool (default: True)

Whether to add the paraview.

spatial_key str (default: 'spatial')

The key in adata.obsm where the spatial coordinates are stored.

set_diag bool (default: False)

Whether to set the diagonal of the connectivity matrix to 1.

kernel Literal['gaussian', 'exponential', 'linear', 'misty_rbf'] (default: 'misty_rbf')

A radial basis function kernel to use for the generation of the connectivity matrix for the paraview. Default is ‘misty_rbf’, a kernel derivative of a Gaussian kernel.

bandwidth float (default: 100)

The bandwidth of the kernel.

zoi float (default: 0)

The zone of indifference of the kernel, i.e. the kernel is set to 0 for distances smaller than zoi.

cutoff float (default: 0.1)

The cutoff for the connectivity matrix.

add_juxta bool (default: True)

Whether to add the juxtaview. The juxtaview is constructed using only the nearest neighbors. A bandwidth of 5 times the bandwidth of the paraview is used to ensure that the nearest neighbors within the radius.

n_neighs int (default: 6)

The number of neighbors to consider when constructing the juxtaview.

max_neighs int (default: 18)

The maximum number of neighbors to consider when constructing the Paraview.

verbose bool (default: False)

Whether to print progress.

Return type:

MistyData

Returns:

MistyData object with the intra view, and two fixed extra view(s): para and juxta.

Examples

Builds the three-view design of the MISTy manuscript from one spatial AnnData: the measurements themselves ('intra'), the immediate neighbours ('juxta'), and the wider neighbourhood ('para'):

>>> import liana as li
>>> adata = li.ds.generate_toy_spatial()
>>> misty = li.mt.genericMistyData(intra=adata, bandwidth=200, set_diag=True)

Pass a second object as extra to predict the intra view from a different modality (e.g. cell-type composition, or pathway activities). Then call the object to fit the model – see liana.mt.MistyData.__call__().