scanpy.pl.pca_loadings

scanpy.pl.pca_loadings(adata, components=None, *, include_lowest=True, n_points=None, show=None, save=None)

Rank genes according to contributions to PCs.

Parameters:

adata AnnData

Annotated data matrix.

components str | Sequence[int] | None (default: None)

For example, '1,2,3' means [1, 2, 3], first, second, third principal component.

include_lowest bool (default: True)

Whether to show the variables with both highest and lowest loadings.

show bool | None (default: None)

Show the plot, do not return axis.

n_points int | None (default: None)

Number of variables to plot for each component.

save str | bool | None (default: None)

If True or a str, save the figure. A string is appended to the default filename. Infer the filetype if ending on {'.pdf', '.png', '.svg'}.

Examples

import scanpy as sc
adata = sc.datasets.pbmc3k_processed()

Show first 3 components loadings

sc.pl.pca_loadings(adata, components = '1,2,3')