dotools_py.pl.correlation

dotools_py.pl.correlation(adata, group_by='batch', figsize=(3, 4), ax=None, palette='RdBu_r', ticks_size=12, path=None, filename='Correlation.svg', show=True, method='pearson', mode='letters', mask=None, square=True, linewidths=0.1, linecolor='black', annot=True, annot_fontsize=15, annot_color='white', annot_kws=None)

Calculate correlation between samples.

Calculates the pearson, spearman or kendall correlation between categorical metadata for all the genes and plot it using a heatmap representation. There are two modes:

• letters: the color of the squares will be white and the correlation values will be colored based on a gradient

• colors: the squares will be colored based on a gradient and the letters will be white.

The gradient is defined based on the provided palette. The input is expected to be log-normalised data.

Parameters:

adata AnnData

Annotated data matrix.

group_by str (default: 'batch')

Name of a categorical column in adata.obs to groupby.

figsize tuple (default: (3, 4))

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.

palette str | list | LinearSegmentedColormap (default: 'RdBu_r')

String denoting matplotlib colormap. If a list of colors is provided a colormap will be generated.

ticks_size int (default: 12)

Size of the X and Y axis ticks.

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

Path to the folder to save the figure.

filename str (default: 'Correlation.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.

method Literal['spearman', 'pearson', 'kendall'] (default: 'pearson')

Method to use to calculate correlation: Pearson, Spearman or Kendall.

mode Literal['colors', 'letters'] (default: 'letters')

Indicate how the correlation is represented.

mask Literal['upper', 'lower'] (default: None)

If set to lower or upper hide the upper or lower triangle of the heatmap, respectively.

square bool (default: True)

If True, set the Axes aspect to “equal” so each cell will be square-shaped.

linewidths float (default: 0.1)

Width of the lines that divide each cell.

linecolor str (default: 'black')

Color of the lines that divide each cell.

annot bool (default: True)

Add text with correlation values to the squares.

annot_fontsize int (default: 15)

Size of the font.

annot_color str (default: 'white')

Color of the font. Will be ignored when mode is letters.

annot_kws dict | None (default: None)

Additional Keyword arguments for matplotlib.axes.Axes.text()

Return type:

Axes | None

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.correlation(adata, 'batch')