spatialtis.NCD_marker

spatialtis.NCD_marker(data, selected_markers=None, importance_cutoff=0.5, layer_key=None, tree_kwargs=None, test_method='mannwhitneyu', pval=0.01, export_key='ncd_marker', **kwargs)

Identify neighbor cells dependent marker

This method tells you the dependency between markers and its neighbor cell type. The dependency is calculated by building a gradiant boosting tree (in here lightgbm) to determine the feature importance. A statistic test and fold change will be calculated for importance markers and its neighbor cells, the fold change is between marker with cell type at / not at the neighborhood.

Parameters

dataAnnData

The AnnData to work with.

importance_cutofffloat, default: 0.5

Threshold to determine the feature markers.

selected_markerslist of str

Select your interested markers.

layer_keystr

The layer in AnnData to perform analysis.

tree_kwargsdict

The keyword arguments that pass to the boosting tree class, (Default: n_jobs=-1, random_state=0).

test_methodstr, default: ‘mannwhitneyu’

which test method to use, anything from scipy.stats.

pvalfloat

The p-value threshold to determine significance.

export_keystr

The key used to store result.

**kwargs

Config for the analysis, for details check spatialtis.abc.AnalysisBase.