scglue.data.aggregate_obs

scglue.data.aggregate_obs(adata, by, X_agg='sum', obs_agg=None, obsm_agg=None, layers_agg=None, separator=',', nan_sparse=False)[source]

Aggregate obs in a given dataset by certain categories

Parameters:

  • adata (AnnData) – Dataset to be aggregated

  • by (str) – Specify a column in adata.obs used for aggregation, must be discrete.

  • X_agg (typing.Optional[str]) – Aggregation function for adata.X, must be one of {"sum", "mean", ``None}``. Setting to None discards the adata.X matrix.

  • obs_agg (typing.Optional[typing.Mapping[str, str]]) – Aggregation methods for adata.obs, indexed by obs columns, must be one of {"sum", "mean", "majority"}, where "sum" and "mean" are for continuous data, and "majority" is for discrete data. Fields not specified will be discarded.

  • obsm_agg (typing.Optional[typing.Mapping[str, str]]) – Aggregation methods for adata.obsm, indexed by obsm keys, must be one of {"sum", "mean"}. Fields not specified will be discarded.

  • layers_agg (typing.Optional[typing.Mapping[str, str]]) – Aggregation methods for adata.layers, indexed by layer keys, must be one of {"sum", "mean"}. Fields not specified will be discarded.

  • separator (str) – Separator between multiple values in the groupby column

  • nan_sparse (bool) – Whether missing entries in sparse matrix indicate nan

Return type:

AnnData

Returns:

aggregated – Aggregated dataset