omicverse.pl.dynamic_trends

omicverse.pl.dynamic_trends(result, genes=None, *, groups=None, datasets=None, compare_features=False, compare_groups=False, add_line=True, add_interval=True, add_point=False, point_color_by=None, point_palette=None, split_time=None, shared_trunk=False, trunk_color='#3C5488', trunk_alpha=0.25, trunk_linewidth=None, line_palette=None, line_palcolor=None, line_style_by=None, line_styles=None, figsize=(3, 3), nrows=None, ncols=None, scatter_size=8, scatter_alpha=0.2, linewidth=2.0, legend=True, legend_outside=True, legend_loc=None, legend_bbox_to_anchor=None, legend_fontsize=None, legend_ncol=1, sharey=False, xlabel='Pseudotime', ylabel='Expression', title=None, title_fontsize=None, add_grid=True, grid_alpha=0.3, grid_linewidth=0.6, show=None, return_axes=False, return_fig=False, ax=None, verbose=True)

Plot GAM-fitted pseudotime trends for one or more genes.

参数:

• result – Result returned by ov.single.dynamic_features().

• genes (default: None) – Gene name or list of gene names to plot. When None, all fitted genes available in result are shown.

• groups (default: None) – Optional fitted group names to include when result contains multiple lineages, cell types, or conditions.

• datasets (default: None) – Backward-compatible alias of groups. Prefer groups in new code.

• compare_features (bool (default: False)) – Whether to compare multiple selected features on the same panel. When enabled for a single fitted group, the default panel title is left blank to avoid generic titles such as adata.

• compare_groups (bool (default: False)) – Whether to compare multiple fitted groups on the same panel. Groups are the fitted series labels returned by ov.single.dynamic_features(), for example lineages, cell types, or condition names.

• add_line (bool (default: True)) – Whether to draw the fitted GAM trend lines.

• add_interval (bool (default: True)) – Whether to draw confidence interval ribbons when available.

• add_point (bool (default: False)) – Whether to overlay observed expression values. This requires that result was computed with ov.single.dynamic_features() using store_raw=True.

• point_color_by (Optional[str] (default: None)) – Optional column in the stored raw point table used to color observed points independently from the fitted line colors. This enables views such as a single global trend line with points colored by State or subtype. Requires add_point=True and ov.single.dynamic_features(..., store_raw=True, raw_obs_keys=[...]).

• point_palette (default: None) – Optional palette used when point_color_by is categorical.

• split_time (Union[float, dict[str, float], None] (default: None)) – Optional pseudotime position used to render branch-aware comparisons. When provided together with compare_groups=True, each group trend is clipped to pseudotime >= split_time. This can be supplied as a scalar applied to every panel, or as a mapping keyed by gene / panel name with an optional 'default' fallback.

• shared_trunk (bool (default: False)) – Whether to draw a shared pre-split trunk when split_time is provided. The trunk is computed as the mean fitted curve across the compared groups for each gene before the split point, giving a generic branch-aware view that can be reused across Monocle, Slingshot, Palantir, CellRank, and other trajectory methods.

• trunk_color (str (default: '#3C5488')) – Color used for the shared trunk line when shared_trunk=True and a single feature is shown on a panel.

• trunk_alpha (float (default: 0.25)) – Alpha transparency used for the shared trunk confidence ribbon.

• trunk_linewidth (Optional[float] (default: None)) – Optional line width used for the shared trunk. Defaults to linewidth when unset.

• line_palette (default: None) – Optional named matplotlib colormap or palette-like sequence describing line colors.

• line_palcolor (default: None) – Explicit color mapping or sequence overriding line_palette.

• line_style_by (Optional[str] (default: None)) – Optional semantic used to vary line styles when a comparison panel contains multiple series. Choose from 'features' or 'groups'.

• line_styles (default: None) – Optional line-style mapping or sequence used with line_style_by.

• figsize (tuple (default: (3, 3))) – Size of each panel in inches. For multi-gene plots this is interpreted as the per-panel size before tiling.

• nrows (Optional[int] (default: None)) – Number of subplot rows for multi-panel layouts.

• ncols (Optional[int] (default: None)) – Number of subplot columns for multi-gene layouts.

• scatter_size (float (default: 8)) – Marker size used for observed points when add_point=True.

• scatter_alpha (float (default: 0.2)) – Alpha transparency for observed points when add_point=True.

• linewidth (float (default: 2.0)) – Line width for fitted trend curves.

• legend (bool (default: True)) – Whether to draw a legend on each axis.

• legend_outside (bool (default: True)) – Whether to place the legend outside the plotting area on the right. Kept for backward compatibility; prefer legend_loc in new code.

• legend_loc (Union[str, int, None] (default: None)) – Legend location. Use 'right margin' for the Scanpy/OmicVerse-style outside-right legend, or any Matplotlib legend location such as 'best' or 'upper center'.

• legend_bbox_to_anchor (Optional[tuple] (default: None)) – Optional Matplotlib bbox_to_anchor used together with legend_loc for precise placement.

• legend_fontsize (Union[float, str, None] (default: None)) – Optional legend font size.

• legend_ncol (int (default: 1)) – Number of legend columns.

• sharey (bool (default: False)) – Whether subplots should share the y-axis in multi-gene layouts.

• xlabel (str (default: 'Pseudotime')) – Label used for the x-axis.

• ylabel (str (default: 'Expression')) – Label used for the y-axis.

• title (default: None) – Optional title applied to the plotted axis or axes. Pass a list/tuple to specify per-panel titles in multi-panel layouts. By default, panel titles use the plotted gene or group label automatically. Pass '' to hide titles explicitly.

• title_fontsize (Optional[float] (default: None)) – Font size used for panel or figure titles. When None, matplotlib’s default title size is used.

• add_grid (bool (default: True)) – Whether to draw a light background grid.

• grid_alpha (float (default: 0.3)) – Alpha transparency used for the background grid.

• grid_linewidth (float (default: 0.6)) – Line width used for the background grid.

• show (Optional[bool] (default: None)) – Whether to call plt.show() before returning. By default, plots are shown only when neither return_fig nor return_axes is requested.

• return_axes (bool (default: False)) – Whether to return the created axis or axes. When False and return_fig=False, the function returns None to avoid notebook repr noise.

• return_fig (bool (default: False)) – Whether to return the figure together with the created axes.

• ax (default: None) – Existing matplotlib axis used when plotting a single gene.

• verbose (bool (default: True)) – Whether to print a short plotting summary.

返回:

Returns None by default after drawing so notebook cells do not print raw axis representations. When return_axes=True, returns the created axis or axes. When return_fig=True, returns (figure, axes).

返回类型:

None | matplotlib.axes.Axes | list[matplotlib.axes.Axes] | tuple