cospar.pp.initialize_adata_object

cospar.pp.initialize_adata_object(adata=None, X_state=None, cell_names=None, gene_names=None, time_info=None, X_clone=None, X_pca=None, X_emb=None, state_info=None, data_des='cospar')

Initialized the AnnData object.

The principal components (X_pca), 2-d embedding (X_emb), and state_info can be provided upfront, or generated in the next step. The clonal information (X_clone) is also optional.

Parameters
adata : AnnData object

X_state : np.ndarray or sp.spmatrix, optional (default: None)

The count matrix for state information. Rows correspond to cells and columns to genes. If adata is provided, this is not necessary.

cell_names : np.ndarray, optional (default: None)

An array of cell ids. If adata is provided, this is not necessary.

gene_names : np.ndarray, optional (default: None)

An array of gene names. If adata is provided, this is not necessary.

time_info : np.ndarray

Time annotation for each cell in str,like ‘Day27’ or ‘D27’. However, it can also contain other sample_info, like ‘GFP+_day27’, and ‘GFP-_day27’.

X_clone : sp.spmatrix or np.ndarray, optional (default: None)

The clonal data matrix, with the row in cell id, and column in barcode id. For evolvable barcoding, a cell may carry several different barcode id. Currently, we assume each entry is 0 or 1.

X_pca : np.ndarray, optional (default: None)

A matrix of the shape n_cell*n_pct. Used for computing similarity matrices.

X_emb : np.ndarray, optional (default: None)

Two-dimensional matrix for embedding.

state_info : np.ndarray, optional (default: None)

The classification and annotation for each cell state. This will be used only after the map is created. Can be adjusted later.

data_des : str, optional (default:’cospar’)

This is just a name to label/distinguish this data. Will be used for saving the results. It should be a unique name for a new dataset stored in the same folder to avoid conflicts.

Returns

  • Generate an AnnData object with the following entries

  • adata.var_names (gene names)

  • adata.obs_names (cell IDs)

  • obs (‘time_info’, ‘state_info’)

  • uns (‘data_des’, ‘clonal_time_points’, ‘time_ordering’)

  • obsm (‘X_clone’, ‘X_pca’, ‘X_emb’)