lynx.trajectory.get_curve

lynx.trajectory.get_curve(adata, use_rep='X_z', n_nodes=20, epg_mu=1.0, epg_lambda=0.1, trim_radius_ratio=0.1, n_repeat=1)[source]

Compute a smooth linear trajectory (t) in [0, 1] via principal graph fitted onto the given manifold (adata.obsm[use_rep]); Use optional marker to rotate the (+/-) sign of the trajectory s.t. the root_marker enriched end is close to t=0.

Parameters:

  • adata (sc.AnnData) – AnnData of latent representation w/ computed elastic principal graph

  • root_marker (str) – Optional marker close to ‘root’ to rotate (+/-) of the trajectory

  • use_rep (str) – Use the indicated representation. ‘X’ or any key for .obsm is valid. If None, the representation is chosen automatically

  • n_nodes (int) – # principal nodes to infer Increase n_nodes get more localized principal manifold

  • trim_radius_ratio (float = 0.1) – Ratio to define trimming radius against embedding value range for robust fitting

  • epg_mu (float)

  • epg_lambda (float)

  • n_repeat (int)