crested.pl.modisco.clustermap_tomtom_similarities#
- crested.pl.modisco.clustermap_tomtom_similarities(sim_matrix, ids, pattern_dict, group_info=None, query_id=None, threshold=None, min_seqlets=0, class_names=None, cmap='viridis', dendrogram_ratio=(0.05, 0.05), logo_width_fraction=0.35, logo_x_padding=0.5, show_pwms=True, save_path=None, width=10, height=10, plot_kws=None, figsize='deprecated')#
Create a Seaborn clustermap of TOMTOM similarity scores with optional PWM logo display and filtering.
- Parameters:
sim_matrix (
ndarray) – 2D square array of TOMTOM similarity scores (-log10 p-values), shape (N, N).ids (
list[str]) – List of pattern identifiers corresponding to rows/columns of sim_matrix.pattern_dict (
dict[str,dict]) – Dictionary mapping pattern IDs to metadata. Each entry should contain: - ‘n_seqlets’: number of seqlets contributing to the pattern. - ‘contrib_scores’: DataFrame or array used for PWM logo plotting.group_info (
list[tuple[list[str],dict[str,str]]] (default:None)) – List of (group_labels, color_map) tuples. Each group_labels list has the same length as ids, and each color_map assigns colors to group values.query_id (
str|None(default:None)) – If provided, only show motifs with similarity >thresholdto this ID.threshold (
float|None(default:None)) – Minimum TOMTOM score for similarity filtering (used only withquery_id).min_seqlets (
int(default:0)) – Minimum number of seqlets required for a pattern to be shown.class_names (
list[str] |None(default:None)) – If provided, only keep patterns whose class name (parsed as ‘_’.join(id.split(‘_’)[:-3])) is in this list.cmap (
str|Colormap(default:'viridis')) – Colormap to use in the clustermap.dendrogram_ratio (tuple[float, float]) – Ratio of dendrogram size to figure size for rows and columns.
logo_width_fraction (
float(default:0.35)) – Width of the PWM logo strip relative to the heatmap width.logo_x_padding (
float(default:0.5)) – Horizontal space between the PWM logos and the heatmap.show_pwms (
bool(default:True)) – Whether to display PWM logos to the left of the heatmap.save_path (
str|None(default:None)) – If provided, the figure is saved to this path (e.g., as a PNG or PDF).plot_kws (
dict|None(default:None)) – Extra keyword arguments passed toclustermap().
- Return type:
ClusterGrid- Returns:
sns.matrix.ClusterGrid The Seaborn clustermap object containing the heatmap and dendrograms.