.. _sphx_glr_auto_examples_03_connectivity_plot_ica.py:


ICA on mouse
============
Independent components analysis on 5 mice.



Retrieve the fMRI data
----------------------



.. code-block:: python

    from sammba import data_fetchers

    retest = data_fetchers.fetch_zurich_test_retest(subjects=range(5),
                                                    correct_headers=True)







Load the template and its brain mask
------------------------------------



.. code-block:: python

    dorr = data_fetchers.fetch_atlas_dorr_2008(downsample='100')
    dorr_masks = data_fetchers.fetch_masks_dorr_2008(downsample='100')
    print('Path to template is {} and to the brain mask is {}'.format(dorr.t2,
          dorr_masks.brain))





.. rst-class:: sphx-glr-script-out

 Out::

    Path to template is /home/salma/nilearn_data/dorr_2008/Dorr_2008_average_100um.nii.gz and to the brain mask is /home/salma/nilearn_data/dorr_2008/dorr_2008_brain_mask_100.nii.gz


Register to the template
------------------------



.. code-block:: python

    import os
    from sammba.registration import TemplateRegistrator

    registrator = TemplateRegistrator(brain_volume=400, caching=True,
                                      template=dorr.t2, use_rats_tool=False,
                                      template_brain_mask=dorr_masks.brain,
                                      registration_kind='affine')

    registered_funcs = []
    for anat, func in zip(retest.anat, retest.func):
        animal_id = os.path.basename(os.path.dirname(anat))
        registrator.output_dir = os.path.join('ica', animal_id)
        registrator.fit_anat(anat)
        registrator.fit_modality(func, 'func', t_r=1., voxel_size=(.3, .3, .3),
                                 prior_rigid_body_registration=True)
        registered_funcs.append(registrator.registered_func_)







Run ICA
-------



.. code-block:: python

    from nilearn.decomposition import CanICA

    canica = CanICA(n_components=30, smoothing_fwhm=.3, n_jobs=-1)
    canica.fit(registered_funcs)







Retrieve the independent components in brain space.



.. code-block:: python

    components_img = canica.masker_.inverse_transform(canica.components_)







Visualize the components
------------------------
We can plot the outline of all components on one figure.



.. code-block:: python

    from nilearn import plotting

    plotting.plot_prob_atlas(components_img,
                             bg_img=registrator.template_brain_,
                             display_mode='z',
                             title='All ICA components')



.. image:: /auto_examples/03_connectivity/images/sphx_glr_plot_ica_001.png
    :align: center




**Total running time of the script:** ( 8 minutes  31.678 seconds)



.. container:: sphx-glr-footer


  .. container:: sphx-glr-download

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



  .. container:: sphx-glr-download

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

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

    `Generated by Sphinx-Gallery <https://sphinx-gallery.readthedocs.io>`_