navis.interfaces.r.nblast_allbyall¶

navis.interfaces.r.nblast_allbyall(x, normalized=True, k=5, resample=None, n_cores=0, use_alpha=False)[source]¶

All-by-all NBLAST using R’s nat.nblast::nblast_allbyall.

NBLAST is optimized for data in microns. Original nat function can be found here.

Parameters:

x (NeuronList | nat.neurons) – (Tree)Neurons to blast. While not strictly necessary, data should be in microns.
k (int, optional) – Number of nearest neighbors to use for dotprops generation. Only relevant if input data is not already Dotprops.
resample (int | bool, optional) – Resampling during dotprops generation. A good value is 1 which means if data is in microns (which it should!) it will be resampled to 1 tangent vector per micron. Only relevant if input data is not already Dotprops.
normalized (bool, optional) – If True, matrix is normalized using z-score.
n_cores (int, optional) – Number of cores to use for nblasting. Default is os.cpu_count() - 2.
use_alpha (bool, optional) – Emphasizes neurons’ straight parts (backbone) over parts that have lots of branches.

Returns:

DataFrame containing the results.

Return type:

pandas.DataFrame

Examples

>>> import navis
>>> from navis.interfaces import r
>>> nl = navis.example_neurons()
>>> # Blast against each other (note the division to get to microns)
>>> scores = r.nblast_allbyall(nl / 1000)