KeyError: 'branch_masks not found in ad.obsm'

[simonekats]

I have computed the PR_results but for some reason it s not adding the branch_probs or branch_masks to the adata. When I check pr_res.branch_probs it shows results but not sure why it is not being added to the adata object. Any advice?

[katosh]

Hi @simonekats, thank you for reporting! Could you provide

• the version of palantir you are uising,
• a code example of the calls you expect to manipulate the anndata,
• and the output of print(ad) before and after the call.

This might allow us to debug this further.

[simonekats]

Hi!

• palantir-1.3.3
• pr_res = palantir.core.run_palantir(
  adata, start_cell, num_waypoints=500, terminal_states = terminal_states
  )
• The output remain the same except it adds the pseudotime and entropy to the adata.obs

[katosh]

That looks great! Please provide the outputs of print(ad) for me to be able to help you though.

[simonekats]

AnnData object with n_obs × n_vars = 457 × 1198
obs: 'experiment_id', 'sample_id', 'count_path', 'Project_ID', 'Experiment_ID', 'Sample_ID', 'Media', 'Condition', 'Maturation', 'n_genes', 'n_genes_by_counts', 'total_counts', 'total_counts_mt', 'pct_counts_mt', 'leiden', 'palantir_pseudotime', 'palantir_entropy', 'macrostates_fwd', 'clusters_gradients', 'term_states_fwd', 'term_states_fwd_probs'
var: 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std', 'gene_trends_clusters', 'mean expression'
uns: 'Condition_colors', 'Experiment_ID_colors', 'Maturation_colors', 'Project_ID_colors', 'hvg', 'leiden', 'leiden_colors', 'log1p', 'neighbors', 'pca', 'tsne', 'umap', 'DM_EigenValues', 'palantir_waypoints', 'masks_key_columns', 'gene_trend_key_branch_pseudotime', 'branch_names', 'gene_trends_M2_1_pseudotime', 'gene_trends_M2_2_pseudotime', 'gene_trends_M2_pseudotime', 'schur_matrix_fwd', 'eigendecomposition_fwd', 'macrostates_fwd_colors', 'coarse_fwd', 'clusters_gradients_colors', 'term_states_fwd_colors', 'T_fwd_params'
obsm: 'X_pca', 'X_tsne', 'X_umap', 'DM_EigenVectors', 'DM_EigenVectors_multiscaled', 'palantir_fate_probabilities', 'branch_masks', 'masks_key', 'schur_vectors_fwd', 'macrostates_fwd_memberships', 'term_states_fwd_memberships', 'lineages_fwd', 'X_fate_simplex_fwd', 'T_fwd_tsne', 'T_fwd_umap'
varm: 'PCs', 'gene_trend_key_branch', 'gene_trends_M2_1', 'gene_trends_M2_2', 'gene_trends_M2', 'terminal_lineage_drivers'
layers: 'MAGIC_imputed_data'
obsp: 'connectivities', 'distances', 'DM_Kernel', 'DM_Similarity'

[katosh]

Its right there: