cospar.tmap.infer_Tmap_from_clonal_info_alone

cospar.tmap.infer_Tmap_from_clonal_info_alone(adata_orig, method='naive', clonal_time_points=None, later_time_point=None, selected_fates=None)

Compute transition map using only the lineage information.

As in infer_Tmap_from_multitime_clones(), we provide two modes of inference:

• If later_time_point=None, the inferred map allows transitions between neighboring time points. For example, if clonal_time_points=[‘day1’,’day2’,’day3’], then it computes transitions for pairs (day1, day2) and (day2, day3), but not (day1, day3).

• If later_time_point is specified, the function produces a map between earlier time points and this later time point. For example, if later_time_point=’day3, the map allows transitions for pairs (day1, day3) and (day2, day3), but not (day1,day2).

Parameters

adata_orig : AnnData object

method : str, optional (default: ‘naive’)

Method used to compute the transition map. Choice: {‘naive’, ‘weinreb’}. For the naive method, we simply average transitions across all clones, assuming that the intra-clone transitions are uniform within the same clone. For the ‘weinreb’ method, we first find uni-potent clones, then compute the transition map by simply averaging across all clonal transitions as the naive method.

selected_fates : list, optional (default: all selected)

List of targeted fate clusters to define uni-potent clones for the weinreb method, which are used to compute the transition map.

clonal_time_points : list of str, optional (default: all time points)

List of time points to be included for analysis. We assume that each selected time point has clonal measurements.

later_time_points : list, optional (default: None)

If specified, the function will produce a map T between these early time points among clonal_time_points and the later_time_point. If not specified, it produces a map T between neighboring time points.

Returns

adata (AnnData object) – The transition map is stored at adata.uns[‘clonal_transition_map’]