AnnData ↭ SingleCellExperiment

RPy2 converter from AnnData_ to SingleCellExperiment_ and back. (For details about conversion see the docs_)

You can for example use it to process your data using both Scanpy_ and Seurat_, as described in this example notebook_

Installation

.. code-block:: bash

pip install anndata2ri

conda install -c bioconda anndata2ri

If you have problems installing or importing anndata2ri,
please make sure you first:

Check the stack trace:
If the error happens while installing or importing a dependency such as rpy2_,
report your problem in that project’s bug tracker.
Search the issues_.
At the time of writing 17 of the 29 bugs (60%) are invalid or rpy2 bugs / install problems.
Make sure you have a compatible R version: rpy2 requires R ≥ 3.6.

Either use the converter manually …

.. code-block:: python

import anndata2ri
from rpy2.robjects import r
from rpy2.robjects.conversion import localconverter

with localconverter(anndata2ri.converter):
adata = r('as(some_data, "SingleCellExperiment")')

… or activate it globally:

.. code-block:: python

import anndata2ri
from rpy2.robjects import r
anndata2ri.activate()

adata = r('as(some_data, "SingleCellExperiment")')

Activate the conversion before you load the extension:

.. code-block:: python

import anndata2ri
anndata2ri.activate()
%load_ext rpy2.ipython

Now you can move objects from Python to R …

.. code-block:: python

import scanpy.datasets as scd
adata_paul = scd.paul15()

.. code-block:: r

%%R -i adata_paul
adata_paul # class: SingleCellExperiment ...

… and back:

.. code-block:: r

%%R -o adata_allen
data(allen, package = 'scRNAseq')
adata_allen <- as(allen, 'SingleCellExperiment')

.. code-block:: python

print(adata_allen) # AnnData object with ...