.. _microns_tut:

MICrONS Datasets
****************

The Allen Institute released two large connecotmics dataset:
    1. A "Cortical mm^3" of mouse visual cortex (broken into two portions, "65" and "35")
    2. A smaller "Layer 2/3" dataset of mouse visual cortex
    
Both of these can be browsed via the `MICrONS Explorer <https://www.microns-explorer.org/>`_ using neuroglancer. These data are public and thanks to the excellent ``cloud-volume`` and the ``caveclient`` libraries (developed by William Silversmith, Forrest Collman, Sven Dorkenwald, Casey Schneider-Mizell and others) we can easily fetch neurons and their connectivity. 

For easier interaction, `navis` ships with a small interface to these datasets. To use it, we will have to make sure ``caveclient`` (and with it ``cloud-volume``) is installed:

.. code-block:: bash

   pip3 install caveclient cloud-volume -U
   
The first time you run below code, you might have to get & set a client secret. Simply follow the instructions in the terminal and when in doubt, check out the section about authentication in the `docs <https://caveclient.readthedocs.io/en/latest/guide/authentication.html>`_.

Let's get started:

.. code:: ipython3

    import navis
    import navis.interfaces.microns as mi

You will find that most functions in the interface accept a ``datastack`` parameter. At the time of writing, the available stacks are:

 - `cortex65` (also called "minnie65") is the anterior portion of the cortical mm^3 dataset
 - `cortex35` (also called "minnie35") is the (smaller) posterior portion of the cortical mm^3 dataset
 - `layer 2/3` (also called "pinky100") is the earlier, smaller cortical dataset 
 
If not specified, the default is `cortex65`! Let's start with some basic queries using ``caveclient``:

.. code:: ipython3

    # Initialize the client for the 65 part of cortical mm^3 (i.e. "Minnie")
    client = mi.get_cave_client(datastack='cortex65')
    
    # Fetch available annotation tables
    client.materialize.get_tables()




.. parsed-literal::

    ['nucleus_detection_v0',
     'synapses_pni_2',
     'nucleus_neuron_svm',
     'proofreading_status_public_release',
     'func_unit_em_match_release',
     'allen_soma_ei_class_model_v1',
     'allen_visp_column_soma_coarse_types_v1']



These are the available public tables which we can use to fetch soma meta data. Let's check out `allen_soma_ei_class_model_v1`:

.. code:: ipython3

    ct = client.materialize.query_table('allen_soma_ei_class_model_v1')
    ct.head()




.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>id</th>
          <th>valid</th>
          <th>classification_system</th>
          <th>cell_type</th>
          <th>pt_supervoxel_id</th>
          <th>pt_root_id</th>
          <th>pt_position</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>485509</td>
          <td>t</td>
          <td>aibs_coarse_excitatory</td>
          <td>excitatory</td>
          <td>103588564537113366</td>
          <td>864691136740606812</td>
          <td>[282608, 103808, 20318]</td>
        </tr>
        <tr>
          <th>1</th>
          <td>113721</td>
          <td>t</td>
          <td>aibs_coarse_excitatory</td>
          <td>excitatory</td>
          <td>79951332685465031</td>
          <td>864691135366988025</td>
          <td>[110208, 153664, 23546]</td>
        </tr>
        <tr>
          <th>2</th>
          <td>263203</td>
          <td>t</td>
          <td>aibs_coarse_excitatory</td>
          <td>excitatory</td>
          <td>87694643458256575</td>
          <td>864691135181741826</td>
          <td>[166512, 174176, 24523]</td>
        </tr>
        <tr>
          <th>3</th>
          <td>456177</td>
          <td>t</td>
          <td>aibs_coarse_excitatory</td>
          <td>excitatory</td>
          <td>102677963354799688</td>
          <td>864691135337690598</td>
          <td>[275616, 135120, 24873]</td>
        </tr>
        <tr>
          <th>4</th>
          <td>364447</td>
          <td>t</td>
          <td>aibs_coarse_excitatory</td>
          <td>excitatory</td>
          <td>94449079618306553</td>
          <td>864691136883828334</td>
          <td>[216064, 166800, 15025]</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    ct.cell_type.unique()




.. parsed-literal::

    array(['excitatory', 'inhibitory'], dtype=object)



Looks like at this point there is only a rather coarse public classification. Nevertheless, let's fetch a couple excitatory and inhibitory neurons:

.. code:: ipython3

    # Fetch proof-reading status
    pr = client.materialize.query_table('proofreading_status_public_release')
    pr.head()




.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>id</th>
          <th>valid</th>
          <th>pt_supervoxel_id</th>
          <th>pt_root_id</th>
          <th>valid_id</th>
          <th>status_dendrite</th>
          <th>status_axon</th>
          <th>pt_position</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>1</td>
          <td>t</td>
          <td>89529934389098311</td>
          <td>864691136296964635</td>
          <td>864691136296964635</td>
          <td>extended</td>
          <td>non</td>
          <td>[179808, 216672, 23361]</td>
        </tr>
        <tr>
          <th>1</th>
          <td>2</td>
          <td>t</td>
          <td>90584228533843146</td>
          <td>864691136311986237</td>
          <td>864691136311986237</td>
          <td>extended</td>
          <td>non</td>
          <td>[187840, 207232, 22680]</td>
        </tr>
        <tr>
          <th>2</th>
          <td>3</td>
          <td>t</td>
          <td>89528353773943370</td>
          <td>864691135355207119</td>
          <td>864691135355207119</td>
          <td>extended</td>
          <td>non</td>
          <td>[180016, 204592, 22798]</td>
        </tr>
        <tr>
          <th>3</th>
          <td>4</td>
          <td>t</td>
          <td>91077153340676495</td>
          <td>864691135355207375</td>
          <td>864691135355207375</td>
          <td>extended</td>
          <td>non</td>
          <td>[191424, 209888, 22845]</td>
        </tr>
        <tr>
          <th>4</th>
          <td>5</td>
          <td>t</td>
          <td>88897234233461709</td>
          <td>864691136422983727</td>
          <td>864691136422983727</td>
          <td>extended</td>
          <td>non</td>
          <td>[175248, 220944, 23561]</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    # Subset to those neurons that have been proof read
    proofread = pr[pr.status_dendrite.isin(['extented', 'clean']) & pr.status_axon.isin(['extented', 'clean'])].pt_root_id.values
    
    ct = ct[ct.pt_root_id.isin(proofread)]
    ct.shape




.. parsed-literal::

    (167, 7)



.. code:: ipython3

    # Pick 20 "root IDs" (sometimes also called segment IDs) each
    inh_ids = ct[ct.cell_type == 'inhibitory'].pt_root_id.values[: 20]
    exc_ids = ct[ct.cell_type == 'excitatory'].pt_root_id.values[: 20]

Next, we will fetch the meshes including the synapses for these neurons:

.. code:: ipython3

    # Fetch those neurons
    inh = mi.fetch_neurons(inh_ids, lod=2, with_synapses=False)
    exc = mi.fetch_neurons(exc_ids, lod=2, with_synapses=False)
    
    # Inspect
    inh












.. raw:: html

    &lt;class 'navis.core.neuronlist.NeuronList'&gt; containing 20 neurons (270.6MiB)<div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>type</th>
          <th>name</th>
          <th>id</th>
          <th>units</th>
          <th>n_vertices</th>
          <th>n_faces</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>navis.MeshNeuron</td>
          <td>None</td>
          <td>864691135994731946</td>
          <td>1 nanometer</td>
          <td>118433</td>
          <td>238637</td>
        </tr>
        <tr>
          <th>1</th>
          <td>navis.MeshNeuron</td>
          <td>None</td>
          <td>864691135974528623</td>
          <td>1 nanometer</td>
          <td>145727</td>
          <td>293750</td>
        </tr>
        <tr>
          <th>...</th>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
        </tr>
        <tr>
          <th>18</th>
          <td>navis.MeshNeuron</td>
          <td>None</td>
          <td>864691136136830589</td>
          <td>1 nanometer</td>
          <td>54995</td>
          <td>111098</td>
        </tr>
        <tr>
          <th>19</th>
          <td>navis.MeshNeuron</td>
          <td>None</td>
          <td>864691135428608048</td>
          <td>1 nanometer</td>
          <td>397265</td>
          <td>799150</td>
        </tr>
      </tbody>
    </table>
    </div>



These neurons are fairly large despite us using a large ``lod`` (level of detail, higher = coarser). For visualization of such large meshes it can be useful to simplify them a little. For this, you need either ``open3d`` (``pip3 install open3d``), ``pymeshlab`` (``pip3 install pymeshlab``) or Blender 3D on your computer.

.. code:: ipython3

    # Reduce face counts to 1/3 of the original
    inh_ds = navis.simplify_mesh(inh, F=1/3)
    exc_ds = navis.simplify_mesh(exc, F=1/3)
    
    # Inspect (note the lower face/vertex counts)
    inh_ds












.. raw:: html

    &lt;class 'navis.core.neuronlist.NeuronList'&gt; containing 20 neurons (69.4MiB)<div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>type</th>
          <th>name</th>
          <th>id</th>
          <th>units</th>
          <th>n_vertices</th>
          <th>n_faces</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>navis.MeshNeuron</td>
          <td>None</td>
          <td>864691135994731946</td>
          <td>1 nanometer</td>
          <td>40887</td>
          <td>79545</td>
        </tr>
        <tr>
          <th>1</th>
          <td>navis.MeshNeuron</td>
          <td>None</td>
          <td>864691135974528623</td>
          <td>1 nanometer</td>
          <td>50585</td>
          <td>97915</td>
        </tr>
        <tr>
          <th>...</th>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
        </tr>
        <tr>
          <th>18</th>
          <td>navis.MeshNeuron</td>
          <td>None</td>
          <td>864691135446864980</td>
          <td>1 nanometer</td>
          <td>108886</td>
          <td>213560</td>
        </tr>
        <tr>
          <th>19</th>
          <td>navis.MeshNeuron</td>
          <td>None</td>
          <td>864691135428608048</td>
          <td>1 nanometer</td>
          <td>136145</td>
          <td>266382</td>
        </tr>
      </tbody>
    </table>
    </div>



Let's visualize the neurons before running analyses

.. code:: ipython3

    # Create some colors: reds for excitatory, blue for inhibitory
    import seaborn as sns
    colors = {n.id: sns.color_palette('Reds', 5)[i] for i, n in enumerate(exc[:5])}
    colors.update({n.id: sns.color_palette('Blues', 5)[i] for i, n in enumerate(inh[:5])})
    
    # Plot the first 5 neurons each
    fig = navis.plot3d([inh_ds[:5], exc_ds[:5]], color=colors)

.. raw:: html
   :file: figures/3d_microns.html
   
|
|
|
|   

Note that some of these neurons look truncated (i.e. extend outside the imaged volume). We need to keep that in mind when running our analyses! 

First, let's check cable length. For this, we need to roughly skeletonize these meshes:

.. code:: ipython3

    inh_sk = navis.skeletonize(inh, parallel=True)
    exc_sk = navis.skeletonize(exc, parallel=True)
    
    # Inspect 
    inh_sk












.. raw:: html

    &lt;class 'navis.core.neuronlist.NeuronList'&gt; containing 20 neurons (57.7MiB)<div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>type</th>
          <th>name</th>
          <th>id</th>
          <th>n_nodes</th>
          <th>n_connectors</th>
          <th>n_branches</th>
          <th>n_leafs</th>
          <th>cable_length</th>
          <th>soma</th>
          <th>units</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>navis.TreeNeuron</td>
          <td>None</td>
          <td>864691135994731946</td>
          <td>19099</td>
          <td>None</td>
          <td>482</td>
          <td>591</td>
          <td>1.064628e+07</td>
          <td>11789</td>
          <td>1 nanometer</td>
        </tr>
        <tr>
          <th>1</th>
          <td>navis.TreeNeuron</td>
          <td>None</td>
          <td>864691135974528623</td>
          <td>24333</td>
          <td>None</td>
          <td>728</td>
          <td>831</td>
          <td>1.349213e+07</td>
          <td>13008</td>
          <td>1 nanometer</td>
        </tr>
        <tr>
          <th>...</th>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
          <td>...</td>
        </tr>
        <tr>
          <th>18</th>
          <td>navis.TreeNeuron</td>
          <td>None</td>
          <td>864691136136830589</td>
          <td>8257</td>
          <td>None</td>
          <td>276</td>
          <td>341</td>
          <td>4.566808e+06</td>
          <td>5777</td>
          <td>1 nanometer</td>
        </tr>
        <tr>
          <th>19</th>
          <td>navis.TreeNeuron</td>
          <td>None</td>
          <td>864691135428608048</td>
          <td>62330</td>
          <td>None</td>
          <td>1109</td>
          <td>1707</td>
          <td>3.432125e+07</td>
          <td>29927</td>
          <td>1 nanometer</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    # Plot one example (zoom in for details)
    fig = navis.plot3d([inh_sk[0], inh[0]], color=[(1, 0, 0), (1, 1, 1, .5)])

.. raw:: html
   :file: figures/3d_microns_skeleton.html
   
|
|
|
|   

That looks reasonable! Now we can compare cable lengths and other stats:

.. code:: ipython3

    # Create a combined DataFrame 
    import pandas as pd 
    
    inh_stats = inh_sk.summary()
    inh_stats['type'] = 'inhibitory'
    
    exc_stats = exc_sk.summary()
    exc_stats['type'] = 'excitatory'
    
    stats = pd.concat([inh_stats, exc_stats], axis=0)
    stats.head()




.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>type</th>
          <th>name</th>
          <th>id</th>
          <th>n_nodes</th>
          <th>n_connectors</th>
          <th>n_branches</th>
          <th>n_leafs</th>
          <th>cable_length</th>
          <th>soma</th>
          <th>units</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>inhibitory</td>
          <td>None</td>
          <td>864691135994731946</td>
          <td>19099</td>
          <td>None</td>
          <td>482</td>
          <td>591</td>
          <td>1.064628e+07</td>
          <td>11789</td>
          <td>1 nanometer</td>
        </tr>
        <tr>
          <th>1</th>
          <td>inhibitory</td>
          <td>None</td>
          <td>864691135974528623</td>
          <td>24333</td>
          <td>None</td>
          <td>728</td>
          <td>831</td>
          <td>1.349213e+07</td>
          <td>13008</td>
          <td>1 nanometer</td>
        </tr>
        <tr>
          <th>2</th>
          <td>inhibitory</td>
          <td>None</td>
          <td>864691135113360025</td>
          <td>31313</td>
          <td>None</td>
          <td>698</td>
          <td>1188</td>
          <td>1.733779e+07</td>
          <td>18923</td>
          <td>1 nanometer</td>
        </tr>
        <tr>
          <th>3</th>
          <td>inhibitory</td>
          <td>None</td>
          <td>864691135700443515</td>
          <td>45083</td>
          <td>None</td>
          <td>2484</td>
          <td>1255</td>
          <td>2.423469e+07</td>
          <td>24681</td>
          <td>1 nanometer</td>
        </tr>
        <tr>
          <th>4</th>
          <td>inhibitory</td>
          <td>None</td>
          <td>864691136521811345</td>
          <td>10173</td>
          <td>None</td>
          <td>824</td>
          <td>352</td>
          <td>5.620049e+06</td>
          <td>6607</td>
          <td>1 nanometer</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    import matplotlib.pyplot as plt
    
    sns.set_color_codes('muted')
    plt.style.use('dark_background')  # comment this out if you have light background
    
    fig, axes = plt.subplots(1, 3, figsize=(12, 5))
    fig.patch.set_color((0, 0, 0, 0)) # Make background transparent
    
    for col, label, ax in zip(['cable_length', 'n_branches', 'n_leafs'],
                              ['cable length  [nm]', '# of branch points', '# of leafs (tips)'],
                              axes):
        sns.boxplot(data=stats, y=col, x='type', ax=ax, palette=['b', 'r'])
        ax.set_ylabel(label)
        ax.set_xlabel('')
        ax.patch.set_color((0, 0, 0, 0)) # Make background transparent
        
    sns.despine(trim=True)
    plt.tight_layout()
    plt.show()



.. image:: microns_tut_files/microns_tut_22_0.png


.. _sholl: 

Let's try something a bit more sophisticated: a Sholl analysis. 

.. note::

   The Sholl analysis function was added in navis ``1.1.0``.

.. code:: ipython3

    # Perform Sholl analysis
    import numpy as np
    inh_sha = navis.sholl_analysis(inh, center='soma', radii=np.linspace(0, 1e6, 200), parallel=True)
    exc_sha = navis.sholl_analysis(exc, center='soma', radii=np.linspace(0, 1e6, 200), parallel=True)
    
    # The results are lists of DataFrames:
    inh_sha[0].head()












.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>intersections</th>
          <th>cable_length</th>
          <th>branch_points</th>
        </tr>
        <tr>
          <th>radius</th>
          <th></th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>5025.125628</th>
          <td>5</td>
          <td>38364.963752</td>
          <td>19</td>
        </tr>
        <tr>
          <th>10050.251256</th>
          <td>2</td>
          <td>200485.343897</td>
          <td>21</td>
        </tr>
        <tr>
          <th>15075.376884</th>
          <td>4</td>
          <td>70893.601194</td>
          <td>18</td>
        </tr>
        <tr>
          <th>20100.502513</th>
          <td>5</td>
          <td>120444.934439</td>
          <td>27</td>
        </tr>
        <tr>
          <th>25125.628141</th>
          <td>8</td>
          <td>231928.170524</td>
          <td>43</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    # Average across inhibitory and excitatory analyses
    inh_sha_mean = inh_sha[0].copy()
    for i in range(1, len(inh_sha)):
        inh_sha_mean += inh_sha[i].values
    inh_sha_mean /= len(inh_sha)
    
    exc_sha_mean = exc_sha[0].copy()
    for i in range(1, len(exc_sha)):
        exc_sha_mean += exc_sha[i].values
    exc_sha_mean /= len(exc_sha)
    
    inh_sha_mean.head()




.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }
    
        .dataframe tbody tr th {
            vertical-align: top;
        }
    
        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>intersections</th>
          <th>cable_length</th>
          <th>branch_points</th>
        </tr>
        <tr>
          <th>radius</th>
          <th></th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>5025.125628</th>
          <td>4.05</td>
          <td>34586.760812</td>
          <td>17.10</td>
        </tr>
        <tr>
          <th>10050.251256</th>
          <td>1.95</td>
          <td>118360.897060</td>
          <td>18.75</td>
        </tr>
        <tr>
          <th>15075.376884</th>
          <td>2.60</td>
          <td>78204.741723</td>
          <td>13.10</td>
        </tr>
        <tr>
          <th>20100.502513</th>
          <td>4.25</td>
          <td>104355.972463</td>
          <td>21.00</td>
        </tr>
        <tr>
          <th>25125.628141</th>
          <td>5.65</td>
          <td>150967.093900</td>
          <td>27.00</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    # Plot
    fig, ax = plt.subplots(figsize=(10, 5))
    
    inh_sha_mean.intersections.plot(c='r', label='inhibitory')
    exc_sha_mean.intersections.plot(c='b', label='excitatory')
    
    ax.set_ylabel('# of intersections')
    ax.legend()
    ax.patch.set_color((0, 0, 0, 0)) # Make background transparent
    fig.patch.set_color((0, 0, 0, 0)) 
    
    plt.tight_layout()
    plt.show()



.. image:: microns_tut_files/microns_tut_26_0.png


See :func:`navis.sholl_analysis` for ways to fine tune the analysis. Last but not least a quick visualization for one of the neurons:

.. code:: ipython3

    from matplotlib.colors import Normalize
    from matplotlib.cm import ScalarMappable
    from mpl_toolkits.axes_grid1 import make_axes_locatable
    
    # Plot one of the inhibitory neurons
    ix = 0
    fig, ax = navis.plot2d(inh[ix], view=('x', 'y'), figsize=(10, 10), c='w', method='2d')
    
    cmap = plt.get_cmap('viridis')
    
    # Plot Sholl circles and color by number of intersections
    center = inh[ix].soma_pos
    norm = Normalize(vmin=0, vmax=(inh_sha[ix].intersections.max() + 1))
    for r in inh_sha[0].index.values:
        ints = inh_sha[ix].loc[r, 'intersections']
        ints_norm = norm(ints)
        color = cmap(ints_norm)
        
        c = plt.Circle(center[:2], r, ec=color, fc='none')
        ax.add_patch(c)
    
    # Add colorbar
    divider = make_axes_locatable(ax)
    cax = divider.append_axes("right", size="5%", pad=0.05)
    plt.colorbar(ScalarMappable(norm=norm, cmap=cmap), cax=cax, label='# of intersections')
    
    plt.show()



.. image:: microns_tut_files/microns_tut_28_0.png


Again: take these analyses with a fistful of salt since we only sample 20 neurons each and it looked like there was a bias towards inhibitory neuron being more likely truncated! However, based on the above analysis inhibitory neurons tend to be larger and branch less compared to excitatory cortical neurons.


Videos
------
Beautiful data like the MICrONS datasets lend themselves to visualizations. For making high quality videos (and renderings) I recommend you check out the tutorial on navis' Blender :ref:`interface <blender3d>`. Here's a little taster:

.. raw:: html
  
   <iframe width="560" height="315" src="https://www.youtube.com/embed/wl3sFG7WQJc" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>