Neighbors.compute_neighbors() got an unexpected keyword argument 'write_knn_indices'

[akhst7]

Firing scv.pp.moments(adata, n_pcs=30, n_neighbors=30) runs into an error.
...

import scvelo as scv
scv.settings.verbosity = 3  
scv.settings.presenter_view = True  
scv.set_figure_params('scvelo')  
adata=scv.dataset(bonemarrow)
scv.pp.filter_and_normalize(adata, min_shared_counts=20, n_top_genes=2000)
scv.pp.moments(adata, n_pcs=30, n_neighbors=30)

Error output

# paste the error output here, if applicable

---

TypeError Traceback (most recent call last)
Cell In[22], line 1
----> 1 scv.pp.moments(adata, n_pcs=30, n_neighbors=30)

File ~/miniconda3/lib/python3.11/site-packages/scvelo/preprocessing/moments.py:63, in moments(data, n_neighbors, n_pcs, mode, method, use_rep, use_highly_variable, copy)
60 normalize_per_cell(adata)
62 if n_neighbors is not None and n_neighbors > get_n_neighs(adata):
---> 63 neighbors(
64 adata,
65 n_neighbors=n_neighbors,
66 use_rep=use_rep,
67 use_highly_variable=use_highly_variable,
68 n_pcs=n_pcs,
69 method=method,
70 )
71 verify_neighbors(adata)
73 if "spliced" not in adata.layers.keys() or "unspliced" not in adata.layers.keys():

File ~/miniconda3/lib/python3.11/site-packages/scvelo/preprocessing/neighbors.py:257, in neighbors(adata, n_neighbors, n_pcs, use_rep, use_highly_variable, knn, random_state, method, metric, metric_kwds, num_threads, copy)
246 neighbors = _get_hnsw_neighbors(
247 adata=adata,
248 use_rep=use_rep,
(...)
254 **metric_kwds,
255 )
256 elif method in ["umap", "gauss", "rapids"]:
--> 257 neighbors = _get_scanpy_neighbors(
258 adata=adata,
259 n_neighbors=n_neighbors,
260 knn=knn,
261 n_pcs=n_pcs,
262 method=method,
263 use_rep=use_rep,
264 random_state=random_state,
265 metric=metric,
266 metric_kwds=metric_kwds,
267 )
268 else:
269 raise ValueError(
270 f"Provided method={method}. Admissible values are 'umap', 'sklearn', "
271 "'hnsw', 'gauss', and 'rapids'."
272 )

File ~/miniconda3/lib/python3.11/site-packages/scvelo/preprocessing/neighbors.py:40, in _get_scanpy_neighbors(adata, **kwargs)
38 warnings.simplefilter("ignore")
39 neighbors = Neighbors(adata)
---> 40 neighbors.compute_neighbors(write_knn_indices=True, **kwargs)
41 logg.switch_verbosity("on", module="scanpy")
43 return neighbors

TypeError: Neighbors.compute_neighbors() got an unexpected keyword argument 'write_knn_indices'

Versions

# paste the ouput of scv.logging.print_versions() here

scvelo==0.3.1 scanpy==1.10.0.dev219+g3ddf1ee1 anndata==0.10.5.post1 loompy==3.0.7 numpy==1.26.4 scipy==1.12.0 matplotlib==3.8.3 sklearn==1.4.1.post1 pandas==2.2.1
ERROR: XMLRPC request failed [code: -32500]
RuntimeError: PyPI no longer supports 'pip search' (or XML-RPC search). Please use https://pypi.org/search (via a browser) instead. See https://warehouse.pypa.io/api-reference/xml-rpc.html#deprecated-methods for more information.

My system is as follows; 

Python 3.11.7
conda 24.1.2

                    'c.          
                 ,xNMM.          -----------------------------
               .OMMMMo           OS: macOS 14.3.1 23D60 x86_64
               OMMM0,            Host: Mac13,2
     .;loddo:' loolloddol;.      Kernel: 23.3.0
   cKMMMMMMMMMMNWMMMMMMMMMM0:    Uptime: 1 hour, 54 mins
 .KMMMMMMMMMMMMMMMMMMMMMMMWd.    Packages: 256 (brew)
 XMMMMMMMMMMMMMMMMMMMMMMMX.      Shell: zsh 5.9
;MMMMMMMMMMMMMMMMMMMMMMMM:       Resolution: 3440x1440
:MMMMMMMMMMMMMMMMMMMMMMMM:       DE: Aqua
.MMMMMMMMMMMMMMMMMMMMMMMMX.      WM: Quartz Compositor
 kMMMMMMMMMMMMMMMMMMMMMMMMWd.    WM Theme: Blue (Light)
 .XMMMMMMMMMMMMMMMMMMMMMMMMMMk   Terminal: iTerm2
  .XMMMMMMMMMMMMMMMMMMMMMMMMK.   Terminal Font: Monaco 10
    kMMMMMMMMMMMMMMMMMMMMMMd     CPU: Apple M1 Ultra
     ;KMMMMMMMWXXWMMMMMMMk.      GPU: Apple M1 Ultra
       .cooc,.    .,coo:.        Memory: 3697MiB / 131072MiB

</details>

[WeilerP]

@akhst7, I suggest you computer the PCA embedding and neighbor graph directly with the corresponding Scanpy functions, i.e.,

sc.pp.pca(adata)
sc.pp.neighbors(adata, n_pcs=30, n_neighbors=30)

and pass n_pcs=None, n_neighbors=None to moments. That will resolve the issue for now.

[akhst7]

Thanks. It worked beautifully.

[SimonE1220]

Unfortunateley same issue here:
computing neighbors

TypeError Traceback (most recent call last)
in <cell line: 1>()
----> 1 scv.tl.velocity(adata, mode="deterministic")

3 frames
/usr/local/lib/python3.10/dist-packages/scvelo/preprocessing/neighbors.py in _get_scanpy_neighbors(adata, **kwargs)
38 warnings.simplefilter("ignore")
39 neighbors = Neighbors(adata)
---> 40 neighbors.compute_neighbors(write_knn_indices=True, **kwargs)
41 logg.switch_verbosity("on", module="scanpy")
42

TypeError: Neighbors.compute_neighbors() got an unexpected keyword argument 'write_knn_indices'

[BoyuJiang95]

I also encountered same error "Neighbors.compute_neighbors() got an unexpected keyword argument 'write_knn_indices'", even I tried:

scv.pp.pca(adata)
scv.pp.neighbors(adata, n_pcs=30, n_neighbors=30)
scv.pp.moments(n_pcs=None, n_neighbors=None)

[YuriPPAN]

I also encountered same error "Neighbors.compute_neighbors() got an unexpected keyword argument 'write_knn_indices'", even I tried:

scv.pp.pca(adata) scv.pp.neighbors(adata, n_pcs=30, n_neighbors=30) scv.pp.moments(n_pcs=None, n_neighbors=None)

Did you solve the problem? I also found the porblem, but it doesn't work.

[yifanfu01]

Hi guys, @YuriPPAN , @BoyuJiang95 , do you mind that should use sc. rather than scv.?
it seems shuttle from scVelo to scanpy.

[Alvis-Jiang]

Hi guys, @YuriPPAN , @BoyuJiang95 , do you mind that should use sc. rather than scv.? it seems shuttle from scVelo to scanpy.

Hi @yifanfu01, thank you! I uesd the sc and now scvelo works well with my data.

Best,
Boyu