svgbit.STDataset

class svgbit.STDataset(count_df: Union[pandas.core.frame.DataFrame, numpy.ndarray, pathlib.Path, str], coordinate_df: Union[pandas.core.frame.DataFrame, numpy.ndarray, pathlib.Path, str], count_transpose: bool = False, coordinate_transpose: bool = False, count_df_kwargs: dict = {}, coordinate_df_kwargs: dict = {}, make_sparse: bool = True, check_duplicate_genes: bool = True, sort_spots: bool = True)

STDataset: A meta class for discribing Spatial Transcriptomics data.

Parameters

count_dfnp.ndarray, pd.DataFrame, str or Path

Expression matrix for Spatial Transcriptomics Data. If str or Path is given, svgbit will try to read file with given path with pandas.

Default shape: (spot * gene)

coordinate_dfnp.ndarray, pd.DataFrame, str or Path

Coordinates for Spatial Transcriptomics Data. If str or Path is given, svgbit will try to read file with given path with pandas.

Default shape: (spot * 2)

count_transposebool, default False

Whether to transpose count matrix.

coordinate_transposebool, default False

Whether to transpose coordinate dataframe.

count_df_kwargsdict, default {}

Keyword arguments pass to pandas.read_csv if str or Path is given to count_df.

coordinate_df_kwargsdict, default {}

Keyword arguments pass to pandas.read_csv if str or Path is given to coordinate_df.

make_sparsebool, default True

Whether to use sparse DataFrame in order to save memory.

check_duplicate_genesbool, default True

Whether to check duplicated gene names.

sort_spotsbool, default True

Whether to sort spots with spots’ name.

Methods

__init__(count_df, coordinate_df[, ...])
acquire_density([cores])	Acquire local Di and global AI value.
acquire_hotspot(**kwargs)	Acquire hotspot matrix.
acquire_weight([k])	Acquire weight for analysis.
find_clusters([n_svgs, n_svg_clusters, ...])	Find SVG clusters.
to_dense()	Convert count_df with sparse values to dense.
to_sparse()	Convert count_df with dense values to sparse.

Attributes

AI	A Series for AI value.
Di	A DataFrame for local Di value.
coordinate_df	Coordinate information.
count_df	Expression matrix.
genes	An Index for genes' names.
hotspot_df	Hotspot matrix.
n_genes	Number of total genes.
n_spots	Number of total spots.
spot_type	A pd.DataFrame for spot type.
spots	An Index for spots' names.
svg_cluster	SVG cluster result.
weight	Weight used by svgbit.
weight_type	What kind of weight is used.