ehrapy.plot.rank_features_groups_stacked_violin#
- ehrapy.plot.rank_features_groups_stacked_violin(adata, groups=None, n_features=None, groupby=None, feature_symbols=None, *, var_names=None, min_logfoldchange=None, key=None, show=None, save=None, return_fig=False, **kwds)[source]#
Plot ranking of genes using stacked_violin plot.
- Parameters:
adata (
AnnData
) – Annotated data matrix.groups (
Union
[str
,Sequence
[str
],None
]) – List of group names.n_features (
Optional
[int
]) – Number of features to show. Is ignored if feature_names is passed.groupby (
Optional
[str
]) – Which key to group the features by.feature_symbols (
Optional
[str
]) – Key for field in .var that stores feature symbols if you do not want to use .var_names displayed in the plot.var_names (
Union
[Sequence
[str
],Mapping
[str
,Sequence
[str
]],None
]) – Feature names.min_logfoldchange (
Optional
[float
]) – Minimum log fold change to consider.key (
Optional
[str
]) – The key of the calculated feature group rankings (default: ‘rank_features_groups’).return_fig (
Optional
[bool
]) – ReturnsStackedViolin
object. Useful for fine-tuning the plot. Takes precedence over show=False.
- Returns:
If return_fig is True, returns a
StackedViolin
object, else if show is false, return axes dict
Examples
>>> import ehrapy as ep >>> adata = ep.data.mimic_2(encoded=True) >>> ep.pp.knn_impute(adata) >>> ep.pp.log_norm(adata, offset=1) >>> ep.pp.neighbors(adata) >>> ep.tl.leiden(adata, resolution=0.15, key_added="leiden_0_5") >>> ep.tl.rank_features_groups(adata, groupby="leiden_0_5") >>> ep.pl.rank_features_groups_stacked_violin(adata, key="rank_features_groups", n_features=5)
- Preview: