crested.pl.patterns.enhancer_design_steps_predictions

crested.pl.patterns.enhancer_design_steps_predictions#

crested.pl.patterns.enhancer_design_steps_predictions(intermediate, target_classes, obs_names, seperate=False, global_ylim='minmax', n_rows=None, n_cols=None, alpha_seperate=1.0, legend_seperate=False, plot_color=(0.3, 0.5, 0.6), show_fliers=False, fig_rescale=1.0, **kwargs)#

Visualize enhancer design prediction score progression.

Parameters:

  • intermediate (list[dict]) – Intermediate output from enhancer design when return_intermediate is True

  • target_classes (str | list[str]) – Target classes that the predictions will be plotted for. All target classes must be in obs_names.

  • obs_names (Index | list[str]) – All class names either in the form of AnnData.obs_names or as a list.

  • seperate (bool (default: False)) – Whether to plot each design enhancer seperately, or all together as a boxplot. Default is False.

  • global_ylim (str (default: 'minmax')) – Used to set global y-axis limits across plots. Can be one of ‘classification’, ‘minmax’ or None. Default is ‘minmax’ ‘classification’ makes the y-axis limits (0, 1). ‘minmax’ makes the y-axis limit minimum and maximum prediction across all the target classes of all designed enhancers If None, each plot has its y-axis limits seperately selected.

  • n_rows (Optional[int] (default: None)) – Number of rows to use when more than one target class is selected.

  • n_cols (Optional[int] (default: None)) – Number of columns to use when more than one target class is selected.

  • alpha_seperate (float (default: 1.0)) – Line alpha for lines when seperate is True. Default is 1.0.

  • legend_seperate (bool (default: False)) – Whether to plot a legend when seperate is True. Default is False.

  • plot_color (str | tuple (default: (0.3, 0.5, 0.6))) – Boxplot color when seperate is False. Default is (0.3, 0.5, 0.6).

  • show_fliers (bool (default: False)) – Whether to show fliers when seperate is False. Default is False.

  • fig_rescale (float (default: 1.0)) – A scalar to scale the figure size up or down. Default is 1.0.