Cell type neighborhood gallery

Cell type neighborhood gallery#

In this notebook, we will visualize a set of cell type neighborhoods to produce visualizations similar to Figure 2 of Yao et al.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from pathlib import Path

from abc_atlas_access.abc_atlas_cache.abc_project_cache import AbcProjectCache

We will interact with the data using the AbcProjectCache. This cache object tracks which data has been downloaded and serves the path to the requsted data on disk. For metadata, the cache can also directly serve a up a Pandas Dataframe. See the getting_started notebook for more details on using the cache including installing it if it has not already been.

The commented section that calls from_local_cache can be used instead when a download of the data already exists on disk or running on CodeOcean with the attached Allen Brain Cell Atlas or similar mounting of the AWS s3 bucket as a directory through s3fs-fuse or similar.

Change the download_base variable to where you have downloaded the data in your system.

download_base = Path('../../abc_download_root')
abc_cache = AbcProjectCache.from_s3_cache(download_base)

# download_base = Path('../../data/abc_atlas') # Path to the already downloaded data or s3fs-fuse mount.
# abc_cache = AbcProjectCache.from_local_cache(download_base)

abc_cache.current_manifest

'releases/20240330/manifest.json'

Read in the list of neighborhood spatial embedding maps

maps = abc_cache.get_metadata_dataframe(directory='WMB-neighborhoods', file_name='dimension_reduction')
maps.set_index('name', inplace=True)
print("number of maps:", len(maps))
maps

number of maps: 8

	label	description	number_of_cells	cluster_group_label
name
HY-EA-Glut-GABA	UMAP20230830-HY-EA-Glut-GABA	Glutamatergic and GABAergic neurons in hypotha...	295448	CCN20230722_HY-EA-Glut-GABA
MB-HB-CB-GABA	UMAP20230830-MB-HB-CB-GABA	GABAergic neurons in midbrain, hindbrain, and ...	186775	CCN20230722_MB-HB-CB-GABA
MB-HB-Glut-Sero-Dopa	UMAP20230830-MB-HB-Glut-Sero-Dopa	Glutamatergic, serotonergic and dopaminergic n...	180669	CCN20230722_MB-HB-Glut-Sero-Dopa
NN-IMN-GC	UMAP20230830-NN-IMN-GC	Non-neuronal cells, immature neurons, granule ...	1418529	CCN20230722_NN-IMN-GC
Pallium-Glut	UMAP20230830-Pallium-Glut	Glutamatergic neurons from cerebral cortex (de...	1494801	CCN20230722_Pallium-Glut
Subpallium-GABA	UMAP20230830-Subpallium-GABA	Telencephalic GABAergic neurons from cerebral ...	611423	CCN20230722_Subpallium-GABA
TH-EPI-Glut	UMAP20230830-TH-EPI-Glut	Glutamatergic neurons from thalamus and epitha...	126171	CCN20230722_TH-EPI-Glut
WholeBrain	UMAP20230830-WholeBrain	All cells in the whole mouse brain (5322 clust...	4042976	CCN20230722_WholeBrain

Read in the extended 10x and MERFISH cell metadata created in the cluster neighborhood tutorial

cell_10x = abc_cache.get_metadata_dataframe(
    directory='WMB-neighborhoods',
    file_name='10x_cell_metadata_with_group_membership',
    dtype={"neurotransmitter": str}
)
cell_10x.set_index('cell_label', inplace=True)
print("number of 10x cells:", len(cell_10x))

10x_cell_metadata_with_group_membership.csv: 100%|█████████████████████████████████████████████████████████████████████| 1.57G/1.57G [01:09<00:00, 22.6MMB/s]

number of 10x cells: 4042976

cell_merfish = abc_cache.get_metadata_dataframe(
    directory='WMB-neighborhoods',
    file_name='merfish_cell_metadata_with_group_membership',
    dtype={"cell_label": str, "neurotransmitter": str}
)
cell_merfish.set_index('cell_label', inplace=True)
print("number of merfish cells:", len(cell_merfish))

merfish_cell_metadata_with_group_membership.csv: 100%|█████████████████████████████████████████████████████████████████| 1.20G/1.20G [00:50<00:00, 24.0MMB/s]

number of merfish cells: 3938808

Create a set of helper functions to read in the neighborhood specific UMAP coordinates and to plot cells in a UMAP or anatomical context and apply them a set of neighborhood

def fetch_coord(group) :
    mlabel = maps.loc[group, 'label']
    coord = abc_cache.get_metadata_dataframe(
        directory='WMB-neighborhoods',
        file_name=mlabel
    )
    coord.set_index('cell_label', inplace=True)
    return coord

def plot_umap(xx, yy, cc=None, val=None, fig_width = 8, fig_height = 8, cmap=None):
    
    fig, ax = plt.subplots()
    fig.set_size_inches(fig_width, fig_height)
    
    if cmap is not None:
        plt.scatter(xx, yy, s=0.5, c=val, marker='.', cmap=cmap)
    elif cc is not None:
        plt.scatter(xx, yy, s=0.5, color=cc, marker='.')
        
    ax.axis('equal')
    ax.set_xlim(-18, 27)
    ax.set_ylim(-18, 27)
    ax.set_xticks([])
    ax.set_yticks([])
    
    return fig, ax

def plot_sections(df, group, feature, blist, fig_width = 20, fig_height = 5):
    
    fig, ax = plt.subplots(1, len(blist))
    fig.set_size_inches(fig_width, fig_height)
    
    for idx, bsl in enumerate(blist):
        
        filtered = df[df['brain_section_label'] == bsl].copy()
        filtered.sort_values(group, ascending=True, inplace=True)
        
        xx = filtered['x']
        yy = filtered['y']

        vv = filtered[feature].copy()
        pred = (filtered[group] == False)
        vv[pred] = '#f5f5f5'
        
        ax[idx].scatter(xx, yy, s=1.0, color=vv, marker=".")
            
        ax[idx].axis('equal')
        ax[idx].set_xlim(0, 11)
        ax[idx].set_ylim(11, 0)
        ax[idx].set_xticks([])
        ax[idx].set_yticks([])
        
        ax[idx].set_title("%s: %s" % (group, bsl))
        
        
    plt.subplots_adjust(wspace=0.01, hspace=0.01)
    plt.show()
    return fig, ax

Neighborhood: Pallium-Glut#

group = 'Pallium-Glut'
coord = fetch_coord(group)
fig, ax = plot_umap(coord['x'], coord['y'], cc=cell_10x.loc[coord.index, 'subclass_color'])

UMAP20230830-Pallium-Glut.csv: 100%|███████████████████████████████████████████████████████████████████████████████████| 84.5M/84.5M [00:03<00:00, 25.0MMB/s]

../_images/4910f5653fe2b2b4eaf35adc44fcc541730e303985c22bfc38d3ab42396c5339.png

blist = ['C57BL6J-638850.52', 'C57BL6J-638850.50', 'C57BL6J-638850.46']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/5826bd111dee2b4bd31ee3701b2739c5c451596dbf55b0fcd54e55c5349daffe.png

blist = ['C57BL6J-638850.43', 'C57BL6J-638850.40', 'C57BL6J-638850.36']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/7a416977a633dee37daf4d6733943a84c9e1300be1b1b002829d005d0b075fac.png

Neighborhood: Subpallium-GABA#

group = 'Subpallium-GABA'
coord = fetch_coord(group)
fig, ax = plot_umap(coord['x'], coord['y'], cc=cell_10x.loc[coord.index, 'subclass_color'])

UMAP20230830-Subpallium-GABA.csv: 100%|████████████████████████████████████████████████████████████████████████████████| 27.7M/27.7M [00:01<00:00, 15.9MMB/s]

../_images/6bd8718432439318c4bbe226f6804cb6fc816254065dff3bfe99dd18c0bfe845.png

blist = ['C57BL6J-638850.52', 'C57BL6J-638850.50', 'C57BL6J-638850.46']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/312e5ef4254486419c019a5a0140680d0885a80e587ca1cb78fc1077c42e11c5.png

blist = ['C57BL6J-638850.43', 'C57BL6J-638850.40', 'C57BL6J-638850.36']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/306d1080806a8f3cf0ec16ac14b2d1e0ba19c4a0878e460c53465aaf31f82e0d.png

Neighborhood: HY-EA-Glut-GABA#

group = 'HY-EA-Glut-GABA'
coord = fetch_coord(group)
fig, ax = plot_umap(coord['x'], coord['y'], cc=cell_10x.loc[coord.index, 'subclass_color'])

../_images/c52ef314efa94fe6b630f0ac2edcc01fe1e7caecb7cfb556b275aa8d338195ed.png

blist = ['C57BL6J-638850.52', 'C57BL6J-638850.50', 'C57BL6J-638850.46']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/cd45db1c197d06b6f131c106a5615d8c1b11aafa3fe29313604fd959fe60e354.png

blist = ['C57BL6J-638850.43', 'C57BL6J-638850.40', 'C57BL6J-638850.36']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/45b0842b0061b8a6ccdd6417379cae5ce990b9a2243331c4fe72ef936eb0858f.png

Neighborhood: TH-EPI-Glut#

group = 'TH-EPI-Glut'
coord = fetch_coord(group)
fig, ax = plot_umap(coord['x'], coord['y'], cc=cell_10x.loc[coord.index, 'subclass_color'])

../_images/23c600bd5bf8c89496b5075a197c8f0e27cd52ccb335e53cf88962c1c30f4e41.png

blist = ['C57BL6J-638850.44', 'C57BL6J-638850.40', 'C57BL6J-638850.38']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/af175172c17782d66ab9c8886a4cd39e602e8ad0f85a04757ccb9fe191ccc044.png

blist = ['C57BL6J-638850.36', 'C57BL6J-638850.33', 'C57BL6J-638850.30']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/13438f40d96ea5213a791bc8b0b6c8dba80bb6f9f53fc3cac930f12795d6b3ac.png

Neighborhood: MB-HB-CB-GABA#

group = 'MB-HB-CB-GABA'
coord = fetch_coord(group)
fig, ax = plot_umap(coord['x'], coord['y'], cc=cell_10x.loc[coord.index, 'subclass_color'])

UMAP20230830-MB-HB-CB-GABA.csv: 100%|██████████████████████████████████████████████████████████████████████████████████| 9.32M/9.32M [00:00<00:00, 18.7MMB/s]

../_images/06e5f1ed2da69e77d21905ce7e19c4f7dcf842b9414fd623b19048fe83e86f08.png

blist = ['C57BL6J-638850.36', 'C57BL6J-638850.30', 'C57BL6J-638850.26']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/b4f869f57f1133381b531dba160374d006acd3f4a5a421338bfe0ace96a198b4.png

blist = ['C57BL6J-638850.16', 'C57BL6J-638850.08', 'C57BL6J-638850.04']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/aae1c57365ff98c44b1ca98525d16cb15ca8cd3d0c8dd47cb865fa4da1ac9ac0.png

Neighborhood: MB-HB-Glut-Sero-Dopa#

group = 'MB-HB-Glut-Sero-Dopa'
coord = fetch_coord(group)
fig, ax = plot_umap(coord['x'], coord['y'], cc=cell_10x.loc[coord.index, 'subclass_color'])

UMAP20230830-MB-HB-Glut-Sero-Dopa.csv: 100%|███████████████████████████████████████████████████████████████████████████| 8.42M/8.42M [00:00<00:00, 18.5MMB/s]

../_images/cf8cdcfe120545151a770e486b0c9d198712c39a486e0528f0468503a159915d.png

blist = ['C57BL6J-638850.36', 'C57BL6J-638850.30', 'C57BL6J-638850.26']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/f30f249ff57aaa1b05312eb1a47d7b972a7b0e0800630aa9f6d5cad363e4b0b1.png

blist = ['C57BL6J-638850.16', 'C57BL6J-638850.08', 'C57BL6J-638850.04']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/af81d9105ffaa1d8900b5928fb0591dd5aed3d3db95d7e84fddcb43fdc1622fa.png

Neighborhood: NN-IMN-GC#

group = 'NN-IMN-GC'
coord = fetch_coord(group)
fig, ax = plot_umap(coord['x'], coord['y'], cc=cell_10x.loc[coord.index, 'subclass_color'])

UMAP20230830-NN-IMN-GC.csv: 100%|██████████████████████████████████████████████████████████████████████████████████████| 83.4M/83.4M [00:03<00:00, 21.4MMB/s]

../_images/9209914dd5fdd0d700f7282e8993c19515030fcc42c7f86660b722ca71d8105f.png

blist = ['C57BL6J-638850.67','C57BL6J-638850.60','C57BL6J-638850.46']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/890a8f145b1c9f89d8e1bdab42c090797cbdd62bcb87bafe9ddc023325a79590.png

blist = ['C57BL6J-638850.36', 'C57BL6J-638850.16', 'C57BL6J-638850.04']
fig, ax = plot_sections(cell_merfish, group, 'subclass_color', blist)

../_images/152175ca17a91088fbc05df0b47678bac957a2ef8831c74f107147958c2775b1.png