from pydantic import (
Field,
ValidationInfo,
computed_field,
field_validator,
model_validator,
BaseModel,
ConfigDict
)
from typing import Optional, List, Dict, Union, Literal, Tuple
class CirclePlotParam(BaseModel):
"""Input schema for LIANA's circle_plot visualization for cell-cell communication networks."""
uns_key: Optional[str] = Field(
default="liana_res",
description="Key in adata.uns that contains the LIANA results."
)
liana_res: Optional[str] = Field(
default=None,
description="DataFrame with LIANA results. If None, uses adata.uns[uns_key]."
)
groupby: Optional[str] = Field(
default=None,
description="Key to be used for grouping or clustering cells."
)
source_key: str = Field(
default="source",
description="Column name of the sender/source cell types in liana_res."
)
target_key: str = Field(
default="target",
description="Column name of the receiver/target cell types in liana_res."
)
score_key: Optional[str] = Field(
default=None,
description="Column name of the score in liana_res. If None, the score is inferred from the method."
)
inverse_score: bool = Field(
default=False,
description="Whether to invert the score. If True, the score will be -log10(score)."
)
top_n: Optional[int] = Field(
default=None,
description="Top N entities to plot."
)
orderby: Optional[str] = Field(
default=None,
description="If top_n is not None, order the interactions by this column."
)
orderby_ascending: Optional[bool] = Field(
default=None,
description="If top_n is not None, specify how to order the interactions."
)
orderby_absolute: bool = Field(
default=False,
description="If top_n is not None, whether to order by the absolute value of the orderby column."
)
filter_fun: str = Field(
default=None,
description="Custom filter function to apply to the results,like: lambda x: x['specificity_rank'] <= 0.05"
)
source_labels: Optional[Union[List[str], str]] = Field(
default=None,
description="List of labels to use as source, the rest are filtered out."
)
target_labels: Optional[Union[List[str], str]] = Field(
default=None,
description="List of labels to use as target, the rest are filtered out."
)
ligand_complex: Optional[Union[List[str], str]] = Field(
default=None,
description="List of ligand complexes to filter the interactions to be plotted."
)
receptor_complex: Optional[Union[List[str], str]] = Field(
default=None,
description="List of receptor complexes to filter the interactions to be plotted."
)
pivot_mode: Literal["counts", "mean"] = Field(
default="counts",
description="The mode of the pivot table: 'counts' for number of connections, 'mean' for mean of score values."
)
mask_mode: Literal["and", "or"] = Field(
default="or",
description="The mode of the mask: 'or' to include source or target, 'and' to include source and target."
)
figure_size: Tuple[float, float] = Field(
default=(5, 5),
description="Figure x,y size."
)
edge_alpha: float = Field(
default=0.5,
description="The transparency of the edges."
)
edge_arrow_size: int = Field(
default=10,
description="The size of the arrow."
)
edge_width_scale: Tuple[float, float] = Field(
default=(1, 5),
description="The scale of the edge width."
)
node_alpha: float = Field(
default=1.0,
description="The transparency of the nodes."
)
node_size_scale: Tuple[float, float] = Field(
default=(100, 400),
description="The scale of the node size."
)
node_label_offset: Tuple[float, float] = Field(
default=(0.1, -0.2),
description="The offset of the node label."
)
node_label_size: int = Field(
default=8,
description="The size of the node label."
)
node_label_alpha: float = Field(
default=0.7,
description="The transparency of the node label."
)
class DotPlotParam(BaseModel):
"""Input schema for LIANA's dotplot visualization for cell-cell communication networks."""
uns_key: str = Field(
default="liana_res",
description="Key in adata.uns that contains the LIANA results."
)
filter_fun: str = Field(
default=None,
description="Custom filter function to apply to the results, like: lambda x: x['specificity_rank'] <= 0.05"
)
colour: str = Field(
...,
description="Column in adata.uns[liana_res] to define the colours of the dots."
)
size: str = Field(
...,
description="Column in adata.uns[liana_res] to define the size of the dots."
)
source_labels: Union[List[str], str] = Field(
default=None,
description="List of labels to use as source, the rest are filtered out."
)
target_labels: Union[List[str], str] = Field(
default=None,
description="List of labels to use as target, the rest are filtered out."
)
top_n: int = Field(
default=None,
description="Top N entities to plot."
)
orderby: str = Field(
default=None,
description="If top_n is not None, order the interactions by this column."
)
orderby_ascending: bool = Field(
default=None,
description="If top_n is not None, specify how to order the interactions."
)
orderby_absolute: bool = Field(
default=False,
description="If top_n is not None, whether to order by the absolute value of the orderby column."
)
ligand_complex: Union[List[str], str] = Field(
default=None,
description="List of ligand complexes to filter the interactions to be plotted."
)
receptor_complex: Union[List[str], str] = Field(
default=None,
description="List of receptor complexes to filter the interactions to be plotted."
)
inverse_colour: bool = Field(
default=False,
description="Whether to -log10 the colour column for plotting."
)
inverse_size: bool = Field(
default=False,
description="Whether to -log10 the size column for plotting."
)
cmap: str = Field(
default="viridis",
description="Colour map to use for plotting."
)
size_range: Tuple[int, int] = Field(
default=(2, 9),
description="Define size range. Tuple of (min, max) integers."
)
figure_size: Tuple[float, float] = Field(
default=(8, 6),
description="Figure x,y size."
)
