Using scATAC-seq dataset to deconvolute 10x Visium dataset

[Jadeapple]

Hello, I want to use scATAC-seq dataset to deconvolute 10x Visium dataset, is it feasible?

When I prepare anndata for the regression model, some warings appeared as follows:
cell2location.models.RegressionModel.setup_anndata(adata=adata_sc,
... batch_key='orig.ident', # 10X reaction / sample / batch
... labels_key='celltype' # cell type, covariate used for constructing signatures
... )
:119: FutureWarning: SparseDataset is deprecated and will be removed in late 2024. It has been replaced by the public classes CSRDataset and CSCDataset.

For instance checks, use isinstance(X, (anndata.experimental.CSRDataset, anndata.experimental.CSCDataset)) instead.

For creation, use anndata.experimental.sparse_dataset(X) instead.

An NVIDIA GPU may be present on this machine, but a CUDA-enabled jaxlib is not installed. Falling back to cpu.

What should I do?

[royassis]

Hey

I found an example in the anndata.experimental.sparse_dataset docstring.

Here is my example code:

import anndata
import h5py
from anndata.experimental import sparse_dataset, read_elem

with h5py.File("somefile.h5ad", "r") as f:
    X = sparse_dataset(f["X"])
    adata = anndata.AnnData(
       layers={"backed": X}, obs=read_elem(f["obs"]), var=read_elem(f["var"])
    )
    print(adata.layers["backed"])

[royassis]

Writing this anndata object to a new file does not keep the X dataset in the new format.