multivelovae.velocity_graph

multivelovae.velocity_graph(adata, key='vae', xkey=None, batch_corrected=False, velocity_offset=False, t_perc=1, **kwargs)

Normalize the velocity matrix and computes velocity graph with scvelo.tl.velocity_graph.

Args:

adata (anndata.AnnData):

Anndata output from VAE inference.

key (str, optional):

Key to find layers. Defaults to ‘vae’.

xkey (str, optional):

Default to use smoothed spliced counts. Defaults to ‘shat’ if batch_corrected, otherwise ‘Ms’.

batch_corrected (bool, optional):

Whether the output was generated with batch correction. Defaults to False.

velocity_offset (bool, optional):

Whether to center the velocities of cells in the first t_perc percentile of latent time. Defaults to False.

t_perc (float, optional):

Percentile of latent time to center velocities. Defaults to 1.

kwargs:

Additional parameters passed to scvelo.tl.velocity_graph.