dotools_py.pl.cell_composition

dotools_py.pl.cell_composition(adata, annot_key, condition_key, batch_key, annot_order=None, condition_order=None, subset_cells=None, figsize=(5, 6), ax=None, title='', title_fontproperties=None, xticks_rotation=None, legend_title='', legend_fontproperties=None, legend_ncols=1, path=None, filename='Proportions.svg', show=True, covariates=None, pval_cutoff=0.05, transform='logit', sep=0.5, bar_width=0.2, linewidth=0.9, add_total_ncell=True, get_props=False, random_state=0, **kwargs)

Stacked barplot showing changes in cell-type proportions.

Generates a stacked barplot to show changes in cell-type proportions between different conditions. Significant changes in cell proportions between conditions will be tested with scanpro <https://github.com/loosolab/scanpro> and will be indicated by a discontinued line. The significant p-value/FDR will be indicated in the legend.

Parameters:

adata AnnData

Annotated data matrix.

annot_key str

Name of a categorical column in adata.obs with the annotation to test for significant differences.

condition_key str

Name of a categorical column in adata.obs to group by.

batch_key str

Name of a categorical column in adata.obs with the batch information.

annot_order list | None (default: None)

Order for the categories in adata.obs[annot_key]

condition_order list | None (default: None)

Order for the categories in adata.obs[condition_key]

subset_cells list | None (default: None)

Only show a subset of groups in adata.obs[annot_key]. The test is applied over all cells.

figsize tuple (default: (5, 6))

Figure size, the format is (width, height).

ax Axes | None (default: None)

Matplotlib axes to use for plotting. If not set, a new figure will be generated.

title str (default: '')

Title for the figure.

title_fontproperties Dict[Literal['size', 'weight'], str | int] (default: None)

Dictionary which should contain ‘size’ and ‘weight’ to define the fontsize and fontweight of the title of the figure.

xticks_rotation int (default: None)

Order for the categories in adata.obs[condition_key].

legend_title str (default: '')

Title for the legend.

legend_fontproperties Dict[Literal['size', 'weight'], str | int] (default: None)

Dictionary which should contain ‘size’ and ‘weight’ to define the fontsize and fontweight of the title of the legend.

legend_ncols int (default: 1)

Number of columns for the legend.

path str | PathLike[str] | Path (default: None)

Path to the folder to save the figure.

filename str (default: 'Proportions.svg')

Name of file to use when saving the figure.

show bool (default: True)

If set to False, returns a dictionary with the matplotlib axes.

covariates list | None (default: None)

Additional covariates for the model. See scanpro.

pval_cutoff float (default: 0.05)

P-val/FDR cutoff.

transform str (default: 'logit')

Method of transformation of proportions.

sep float (default: 0.5)

Separation between bars.

bar_width float (default: 0.2)

Bars width.

linewidth float (default: 0.9)

Thickness of the lines connecting significant bars.

add_total_ncell bool (default: True)

Add the total number of cells in the dataset.

get_props bool (default: False)

If set to True, returns a dataframe with the cell proportions.

random_state int (default: 0)

seed for random number generator.

kwargs

Additional arguments pass to scanpro().

Return type:

None | DataFrame | Axes

Returns:

Depending on show, returns the plot if set to True or a dictionary with the axes.

Example

import dotools_py as do
adata = do.dt.example_10x_processed()
do.pl.cell_composition(adata, "annotation", "condition", "batch", condition_order=["healthy", "disease"], transform="arcsin")