drvi.utils.plotting.plot_interpretability_scores

drvi.utils.plotting.plot_interpretability_scores#

drvi.utils.plotting.plot_interpretability_scores(gene_scores_df, n_top_genes=10, ncols=5, score_threshold=0.1, dim_subset=None, show=True, **kwargs)[source]#

Plot bar plots of the top genes per latent dimension for an interpretability score.

This function visualizes the result of get_interpretability_scores(). For each latent dimension (a column of gene_scores_df) it draws a horizontal bar plot of the n_top_genes genes with the highest scores.

Parameters:

  • gene_scores_df (DataFrame) – DataFrame of interpretability scores with genes as rows and latent dimensions as columns, as returned by get_interpretability_scores.

  • n_top_genes (int (default: 10)) – Number of top genes to display per dimension.

  • ncols (int (default: 5)) – Number of columns in the subplot grid.

  • score_threshold (float (default: 0.1)) – Minimum score for a dimension to be plotted. Dimensions whose maximum gene score is below this threshold are skipped.

  • dim_subset (Sequence[str] | None (default: None)) – Subset of dimension titles (column names) to plot. If None, all dimensions meeting score_threshold are plotted, ordered by their dimension number.

  • show (bool (default: True)) – Whether to display the plot. If False, returns the figure object.

  • **kwargs – Additional keyword arguments passed to matplotlib.axes.Axes.barh().

Returns:

matplotlib.figure.Figure or None The figure object if show=False, otherwise None.