pandora.plotting module

pandora.plotting.get_distinct_colors(n_colors: int) → List[str]

Returns a list of n HSV colors evenly spaced in the HSV colorspace.

Parameters:

n_colorsint

Number of colors to return

Returns:

List[str]

List of n plotly HSV color strings.

pandora.plotting.get_rdylgr_color_scale() → List[Tuple[float, str]]

Returns a continuous hex color scale from red (#d60000) to green (#209c05).

Returns:

List[Tuple[float, str]]

A list of (float, str) tuples that can be used as continuous color scale in plotly figures.

pandora.plotting.improve_plotly_text_position(x_values: Series) → List[str]

Returns improved text positions for sample annotations in plotly figures based on the x-values of the samples.

Parameters:

x_valuespd.Series

x values of the samples to plot

Returns:

List[str]

A list of text positions, one position for each sample in x_values.

pandora.plotting.plot_clusters(embedding: Embedding, dim_x: int = 0, dim_y: int = 1, kmeans_k: int | None = None, fig: Figure | None = None, **kwargs) → Figure

Plots the data for the provided Embedding data using the given dimension indices and color-codes samples according to their cluster label as inferred by using K-Means clustering.

Parameters:

embeddingEmbedding

Embedding data to plot.

dim_xint

Index of the dimension plotted on the x-axis (zero-indexed).

dim_yint

Index of the dimension plotted on the y-axis (zero-indexed).

kmeans_kint

Optional k to use for K-Means clustering. If not set, the optimal number of clusters k is automatically determined based on the data provided by Embedding.

figgo.Figure

Optional figure containing previous plotting data (e.g. another Embedding plot).

**kwargs

Optional plot arguments passed to go.Scatter. Refer to the plotly documentation for options. The following settings are not allowed: [x, y, mode, marker, marker_color, name]

Returns:

go.Figure

Plotly figure depicting the Embedding data.

pandora.plotting.plot_embedding_comparison(embedding_comparison: EmbeddingComparison, dim_x: int = 0, dim_y: int = 1, **kwargs) → Figure

Method to plot the closest match between two Embeddings. Plots the transformed Embeddings based on the EmbeddingComparison object.

Parameters:

embedding_comparisonEmbeddingComparison

EmbeddingComparison object containing the two Embeddings to plot.

dim_xint

Index of the dimension plotted on the x-axis (zero-indexed).

dim_yint

Index of the dimension plotted on the y-axis (zero-indexed).

**kwargs

Optional plot arguments passed to go.Scatter. Refer to the plotly documentation for options. The following settings are not allowed: [x, y, mode, marker, marker_color, name, marker_symbol]

Returns:

go.Figure

Plotly figure depicting both Embeddings in embedding_comparison.

pandora.plotting.plot_embedding_comparison_rogue_samples(embedding_comparison: EmbeddingComparison, support_values: Series, support_value_rogue_cutoff: float = 0.5, dim_x: int = 0, dim_y: int = 1, **kwargs) → Figure

Method to plot the closest match between two Embeddings. Plots the transformed Embeddings based on the EmbeddingComparison object. Additionally, all samples with a support value below support_value_rogue_cutoff are highlighted.

Parameters:

embedding_comparisonEmbeddingComparison

EmbeddingComparison object containing the two Embeddings to plot.

support_valuespd.Series

Pandora support values for all samples in the embedding data. The series’ indices should be the sample IDs and the values the respective support value.

support_value_rogue_cutofffloat

Samples with a support value below this threshold are considered rogue and are highlighted by color, their sample ID and support value.

dim_xint

Index of the dimension plotted on the x-axis (zero-indexed).

dim_yint

Index of the dimension plotted on the y-axis (zero-indexed).

**kwargs

Optional plot arguments passed to go.Scatter. Refer to the plotly documentation for options. The following settings are not allowed: [x, y, mode, marker, marker_color, name, ` text`, textposition, marker_symbol]

Returns:

go.Figure

Plotly figure depicting both Embeddings in embedding_comparison and highlighting rouge samples.

pandora.plotting.plot_populations(embedding: Embedding, dim_x: int = 0, dim_y: int = 1, fig: Figure | None = None, **kwargs) → Figure

Plots the data for the provided Embedding data using the given dimension indices and colors all populations as provided by Embedding using distinct colors.

Parameters:

embeddingEmbedding

Embedding data to plot.

dim_xint

Index of the dimension plotted on the x-axis (zero-indexed).

dim_yint

Index of the dimension plotted on the y-axis (zero-indexed).

figgo.Figure, default=None

Optional figure containing previous plotting data (e.g. another Embedding plot).

**kwargs

Optional plot arguments passed to go.Scatter. Refer to the plotly documentation for options. The following settings are not allowed: [x, y, mode, marker, marker_color, name]

Returns:

go.Figure

Plotly figure depicting the Embedding data

Raises:

PandoraException

If there are no populations associated with the embedding data.

pandora.plotting.plot_projections(embedding: Embedding, embedding_populations: Series, dim_x: int = 0, dim_y: int = 1, fig: Figure | None = None, **kwargs)

Plots the data for the provided Embedding data using the given dimension indices. Only samples with populations not in embedding_populations are color-coded according to their population. All other samples are colored in lightgray.

Use this plotting function if you want to highlight only projected samples in an Embedding plot.

Parameters:

embeddingEmbedding

Embedding data to plot.

embedding_populationspd.Series

Pandas Series of population names used to compute the Embedding. Samples belonging to these populations are plotted in lightgray.

dim_xint

Index of the dimension plotted on the x-axis (zero-indexed).

dim_yint

Index of the dimension plotted on the y-axis (zero-indexed).

figgo.Figure

Optional figure containing previous plotting data (e.g. another Embedding plot).

**kwargs

Optional plot arguments passed to go.Scatter. Refer to the plotly documentation for options. The following settings are not allowed: [x, y, mode, marker, marker_color, name]

Returns:

go.Figure

Plotly figure depicting the Embedding data.

Raises:

PandoraException

If embedding_populations is empty. In this case there are not projected samples to plot.

pandora.plotting.plot_support_values(embedding: Embedding, sample_support_values: Series, support_value_rogue_cutoff: float = 0.5, dim_x: int = 0, dim_y: int = 1, projected_samples: Series | None = None, **kwargs) → Figure

Plots the data for the provided Embedding data using the given dimension indices, color-coding the support value for each sample.

The colors range from red (low support) to green (high support). If projected_samples is set, only samples in projected_samples are color-coded according to their support value, all other samples are shown in lightgray.

Parameters:

embeddingEmbedding

Embedding data to plot.

sample_support_valuespd.Series

Pandora support value for each sample in embedding.embedding. Note: The index of the Series is expected to contain the sample IDs and to be identical to embedding.sample_ids.

support_value_rogue_cutofffloat

Samples with a support value below this threshold are annotated with the sample ID and the support value. All other samples are only color-coded.

dim_xint

Index of the dimension plotted on the x-axis (zero-indexed).

dim_yint

Index of the dimension plotted on the y-axis (zero-indexed).

projected_samplespd.Series

Pandas series of sample IDs. If set, only samples in this list are color-coded according to their support value. All other samples are shown in gray.

**kwargs

Optional plot arguments passed to go.Scatter. Refer to the plotly documentation for options. The following settings are not allowed: [x, y, mode, marker, marker_color, name, text, textposition]

Returns:

go.Figure

Plotly figure depicting the Embedding data.

Raises:

PandoraException

If no samples are left to plot after filtering samples present in both embedding data and sample_support_values.

Warns:

UserWarning

• If not all samples in sample_support_values are present in the embedding data. This is most likely due to outlier detection during the computation of the embedding.

• If not all samples in the embedding data have a support value associated.