.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "auto_examples/tools/plot_reclustering_function.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        Click :ref:`here <sphx_glr_download_auto_examples_tools_plot_reclustering_function.py>`
        to download the full example code

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_auto_examples_tools_plot_reclustering_function.py:


reclustering on specific leiden clusters
=========================================

This example demonstrates who to perform a reclustering on a selected subset of
leiden clusters. You will want to do this for example during the process of celltype
annotation, when the  clusters do not have a sufficient resolution to seperate
all clusters and mixed cell populations still exist.

.. GENERATED FROM PYTHON SOURCE LINES 11-71



.. image-sg:: /auto_examples/tools/images/sphx_glr_plot_reclustering_function_001.png
   :alt: leiden, CD3G, CD8A, CD4, IL7R, NKG7, GNLY
   :srcset: /auto_examples/tools/images/sphx_glr_plot_reclustering_function_001.png
   :class: sphx-glr-single-img





.. code-block:: default


    import besca as bc
    import scanpy as sc
    import pytest

    # pytest.skip('Test is only for here as example and should not be executed')
    # load and preprocess data (here we will start from a preprocessed dataset)
    adata = bc.datasets.pbmc3k_processed()

    # extract subset using the recluster function whcih is part of the reclustering (rc) toolkit
    adata_subset = bc.tl.rc.recluster(
        adata,
        celltype=("2", "3", "4", "5", "6", "8", "9", "10", "11", "12"),
        celltype_label="leiden",
        resolution=1.2,
    )


    # visualize the new clusters
    sc.pl.umap(
        adata_subset, color=["leiden", "CD3G", "CD8A", "CD4", "IL7R", "NKG7", "GNLY"]
    )

    # append new celltype labels to the subclusters.
    # This is an approximative hand annotation that should be dealt into more widths.
    labels = [
        "NK cell",  # 0
        "CD4 T-cell",  # 1
        "CD8 T-cell",  # 2
        "CD4 T-cell",  # 3
        "CD8 T-cell",  # 4
        "CD8 T-cell",  # 5
        "CD4 T-cell",  # 6
        "CD4 T-cell",  # 7
        "CD4 T-cell",  # 8
        "CD4 T-cell",  # 9
        "CD4 T-cell",  # 10
        "CD4 T-cell",  # 11
        "CD4 T-cell",  # 12
        "CD4 T-cell",  # 13
        "CD4 T-cell",  # 14
        "CD4 T-cell",  # 15
        "CD4 T-cell",  # 16
        "CD4 T-cell",  # 17
        "CD4 T-cell",  # 18
        "CD4 T-cell",  # 19
    ]  # 10

    new_labels = [
        labels[i]
        for i in range(len(adata_subset.obs.get("leiden").value_counts().index.tolist()))
    ]


    # merge the labels back into the original adata object
    # note this will overwrite what ever was saved in adata.obs.celltype;
    # Here is was not assigned yet.
    bc.tl.rc.annotate_new_cellnames(
        adata, adata_subset, names=new_labels, new_label="celltype"
    )


.. rst-class:: sphx-glr-timing

   **Total running time of the script:** ( 0 minutes  3.241 seconds)


.. _sphx_glr_download_auto_examples_tools_plot_reclustering_function.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example


    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: plot_reclustering_function.py <plot_reclustering_function.py>`

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: plot_reclustering_function.ipynb <plot_reclustering_function.ipynb>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_