Ingest won't integrate datasets of different lengths

• [x ] I have checked that this issue has not already been reported.
• [ x] I have confirmed this bug exists on the latest version of scanpy.
• (optional) I have confirmed this bug exists on the master branch of scanpy.

I am trying to ingest a CITEseq dataset into another clustered dataset. These datasets have different numbers of cells but I ran neighbors(n_neighbors=30) for both prior to running umap. I have confirmed that both datasets have the same variable names and the same number of variable names (38). Both objects look identical when a call adata.var.

I receive the error: "all input arrays must have the same shape".

sc.pp.neighbors(CODEX_sub, n_neighbors=30) 
sc.tl.umap(CODEX_sub)
sc.pp.neighbors(adata_sub, n_neighbors = 30)
sc.tl.umap(adata_sub)
sc.tl.ingest(CODEX_sub, adata_sub, obs='leiden', embedding_method='umap')

ValueError                                Traceback (most recent call last)
<ipython-input-214-01a03312d3df> in <module>
----> 1 sc.tl.ingest(CODEX_sub, adata_sub, obs='leiden', embedding_method='umap')

~\anaconda3\envs\scenv\lib\site-packages\scanpy\tools\_ingest.py in ingest(adata, adata_ref, obs, embedding_method, labeling_method, neighbors_key, inplace, **kwargs)
    124         labeling_method = labeling_method * len(obs)
    125 
--> 126     ing = Ingest(adata_ref, neighbors_key)
    127     ing.fit(adata)
    128 

~\anaconda3\envs\scenv\lib\site-packages\scanpy\tools\_ingest.py in __init__(self, adata, neighbors_key)
    383 
    384         if neighbors_key in adata.uns:
--> 385             self._init_neighbors(adata, neighbors_key)
    386         else:
    387             raise ValueError(

~\anaconda3\envs\scenv\lib\site-packages\scanpy\tools\_ingest.py in _init_neighbors(self, adata, neighbors_key)
    349         else:
    350             self._neigh_random_state = neighbors['params'].get('random_state', 0)
--> 351             self._init_pynndescent(neighbors['distances'])
    352 
    353     def _init_pca(self, adata):

~\anaconda3\envs\scenv\lib\site-packages\scanpy\tools\_ingest.py in _init_pynndescent(self, distances)
    284 
    285         first_col = np.arange(distances.shape[0])[:, None]
--> 286         init_indices = np.hstack((first_col, np.stack(distances.tolil().rows)))
    287 
    288         self._nnd_idx = NNDescent(

<__array_function__ internals> in stack(*args, **kwargs)

~\anaconda3\envs\scenv\lib\site-packages\numpy\core\shape_base.py in stack(arrays, axis, out)
    424     shapes = {arr.shape for arr in arrays}
    425     if len(shapes) != 1:
--> 426         raise ValueError('all input arrays must have the same shape')
    427 
    428     result_ndim = arrays[0].ndim + 1

ValueError: all input arrays must have the same shape

Versions

scanpy==1.7.0 anndata==0.7.6 umap==0.5.1 numpy==1.21.1 scipy==1.7.0 pandas==1.2.5 scikit-learn==0.24.2 statsmodels==0.12.2 python-igraph==0.9.8

Reproduce the failure with the shown sc.tl.ingest sequence, then inspect scanpy/tools/_ingest.py, especially Ingest._init_neighbors and _init_pynndescent where the traceback reaches NumPy stack. Compare the reference neighbor data for datasets with different cell counts and verify that the ingest call completes without the shape error.