Question about 'tangram_ct_pred' results for final cell type annotation call

[KunHHE]

Hi @ziqlu0722, for Tangram run finished and got 'tangram_ct_pred', I don't feel confident for the downstream analysis step; can you help me a little? Thanks!

For example, should I do normalization for the probability value and then project to cell types? like using the scripts below:

probabilities = np.array(comb_adata.obsm['tangram_pred'])
n_voxels = probabilities.shape[0]
n_cell_types = probabilities.shape[1]

predicted_cell_types = [XXXX cell types
]
assert len(predicted_cell_types) == n_cell_types, f"Mismatch: {len(predicted_cell_types)} vs {n_cell_types}"
sampled_cell_types = []
for voxel_idx in range(n_voxels):
voxel_probabilities = probabilities[voxel_idx, :]
voxel_probabilities /= np.sum(voxel_probabilities)
sampled_cell_type_idx = np.random.choice(n_cell_types, p=voxel_probabilities)
sampled_cell_type = predicted_cell_types[sampled_cell_type_idx]
sampled_cell_types.append(sampled_cell_type)
comb_adata.obs['sampled_cell_type'] = sampled_cell_types