sc.external.tl.palantir uses invalid keywork in run_diffusion_maps()

[soerenab]

Please make sure these conditions are met

• I have checked that this issue has not already been reported.
• 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.

What happened?

Running sc.external.tl.palantir(adata) raises TypeError: run_diffusion_maps() got an unexpected keyword argument 'data_df'. This seems to be a bug in scanpy/external/tl/_palantir.py which uses the keyword data_df which seems to have changed from data_df to data in the latest palantir version.

Minimal code sample

# from the documentation: https://scanpy.readthedocs.io/en/stable/generated/scanpy.external.tl.palantir.html#scanpy.external.tl.palantir
# (the data comes with the palantir repo)
import scanpy.external as sce
import scanpy as sc
adata = sc.read_csv(filename="Palantir/data/marrow_sample_scseq_counts.csv.gz")
sc.pp.filter_cells(adata, min_counts=1000)
sc.pp.filter_genes(adata, min_counts=10)
sc.pp.normalize_per_cell(adata)
sc.pp.log1p(adata)
sc.tl.pca(adata, n_comps=300)
sc.pp.neighbors(adata, knn=30)
sce.tl.palantir(adata, n_components=5, knn=30) # error occurs here

Error output

RuntimeError                              Traceback (most recent call last)
RuntimeError: module compiled against API version 0xf but this version of numpy is 0xe
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
/tmp/ipykernel_5717/3469234522.py in <cell line: 7>()
      5 sc.tl.pca(adata, n_comps=300)
      6 sc.pp.neighbors(adata, knn=30)
----> 7 sce.tl.palantir(adata, n_components=5, knn=30)

/p/project/hai_microbio/sb/miniconda3/envs/cellrank2/lib/python3.9/site-packages/scanpy/external/tl/_palantir.py in palantir(adata, n_components, knn, alpha, use_adjacency_matrix, distances_key, n_eigs, impute_data, n_steps, copy)
    207 
    208     # Diffusion maps
--> 209     dm_res = run_diffusion_maps(
    210         data_df=df,
    211         n_components=n_components,

TypeError: run_diffusion_maps() got an unexpected keyword argument 'data_df'

Versions

-----
anndata     0.9.2
scanpy      1.9.3
-----
PIL                 10.0.0
cffi                1.15.1
colorama            0.4.6
cycler              0.10.0
cython_runtime      NA
dateutil            2.8.2
gmpy2               2.1.2
h5py                3.9.0
igraph              0.10.6
importlib_resources NA
joblib              1.3.2
kiwisolver          1.4.4
leidenalg           0.10.1
llvmlite            0.40.1
matplotlib          3.7.2
mpl_toolkits        NA
mpmath              1.3.0
natsort             8.4.0
numba               0.57.1
numexpr             2.8.5
numpy               1.24.4
opt_einsum          v3.3.0
packaging           23.1
pandas              1.5.3
psutil              5.9.5
pyparsing           3.0.9
pytz                2023.3
scipy               1.11.1
session_info        1.0.0
six                 1.16.0
sklearn             1.3.0
sympy               1.12
texttable           1.6.7
threadpoolctl       3.2.0
torch               2.0.0
tqdm                4.65.2
typing_extensions   NA
wcwidth             0.2.6
yaml                6.0
zipp                NA
zoneinfo            NA
-----
Python 3.9.16 | packaged by conda-forge | (main, Feb  1 2023, 21:39:03) [GCC 11.3.0]
Linux-4.18.0-477.15.1.el8_8.x86_64-x86_64-with-glibc2.28
-----
Session information updated at 2023-08-09 17:08

[soerenab]

After fixing the above error locally and continuing I ran into a similar error in the next step:
sc.external.tl.palantir_results(...) internally calls run_palantir(ms_data=ms_data, ...) with keyword ms_data which seems to have changed to just data in the recent version of palantir.

[Zethson]

Would you be willing to file a PR?
In any case, we're deprecating external. Please use Palantir directly.

[katosh]

Palantir can now take anndata directly and the argument names have changed to a more general data. How is the status on this @soerenab? I would gladly provide a PR.