Download Example Tabula Sapiens Anndata Object and Manipulation

!wget https://figshare.com/ndownloader/files/34701991 # not working with Anaconda terminal, has to be wsl terminal

The following block will download the TS_lung.h5ad file into current directory

import requests
from zipfile import ZipFile
from io import BytesIO

# Downloading the file using requests
url = "https://figshare.com/ndownloader/files/34701991"
response = requests.get(url)
response.raise_for_status()

# Unzipping the file using zipfile
with ZipFile(BytesIO(response.content)) as z:
    z.extractall()

import scanpy as sc
adata = sc.read_h5ad("TS_lung.h5ad")
adata

adata.obs

	organ_tissue	method	donor	anatomical_information	n_counts_UMIs	n_genes	cell_ontology_class	free_annotation	manually_annotated	compartment	gender
cell_id
AAACCCAAGAACTCCT_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	44221.0	6970	type ii pneumocyte	type ii pneumocyte	True	epithelial	male
AAACGAAAGGAGCTGT_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	28563.0	5638	type ii pneumocyte	type ii pneumocyte	True	epithelial	male
AAACGAACACGCGTGT_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	4099.0	1929	type ii pneumocyte	type ii pneumocyte	True	epithelial	male
AAACGAACATGCCGCA_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	24619.0	4021	type ii pneumocyte	type ii pneumocyte	True	epithelial	male
AAAGAACGTTAAGGGC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	28129.0	5278	type ii pneumocyte	type ii pneumocyte	True	epithelial	male
…	…	…	…	…	…	…	…	…	…	…	…
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P1_S361	Lung	smartseq2	TSP2	proxmedialdistal	1788070.0	1325	macrophage	macrophage	True	immune	female
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P2_S362	Lung	smartseq2	TSP2	proxmedialdistal	2897053.0	3320	plasma cell	plasma cell	True	immune	female
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P6_S366	Lung	smartseq2	TSP2	proxmedialdistal	7151.0	617	lung microvascular endothelial cell	lung microvascular endothelial cell	True	endothelial	female
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P8_S368	Lung	smartseq2	TSP2	proxmedialdistal	14444.0	404	neutrophil	neutrophil	True	immune	female
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P9_S369	Lung	smartseq2	TSP2	proxmedialdistal	689104.0	6964	type ii pneumocyte	type ii pneumocyte	True	epithelial	female

35682 rows × 11 columns

adata.X

<35682x58870 sparse matrix of type '<class 'numpy.float32'>'
    with 119787345 stored elements in Compressed Sparse Row format>

adata.var_names

Index(['DDX11L1', 'WASH7P', 'MIR6859-1', 'MIR1302-2HG', 'MIR1302-2', 'FAM138A',
       'OR4G4P', 'OR4G11P', 'OR4F5', 'AL627309.1',
       ...
       'MT-ND4', 'MT-TH', 'MT-TS2', 'MT-TL2', 'MT-ND5', 'MT-ND6', 'MT-TE',
       'MT-CYB', 'MT-TT', 'MT-TP'],
      dtype='object', length=58870)

adata.obs.cell_ontology_class.unique().tolist()[0:10]

['type ii pneumocyte',
 'neutrophil',
 'cd4-positive alpha-beta t cell',
 'cd8-positive alpha-beta t cell',
 'nk cell',
 'bronchial vessel endothelial cell',
 'smooth muscle cell',
 'adventitial cell',
 'macrophage',
 'respiratory mucous cell']

bdata = adata[adata.obs.cell_ontology_class == 'neutrophil']

adata.obs.head(1)

	organ_tissue	method	donor	anatomical_information	n_counts_UMIs	n_genes	cell_ontology_class	free_annotation	manually_annotated	compartment	gender
cell_id
AAACCCAAGAACTCCT_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	44221.0	6970	type ii pneumocyte	type ii pneumocyte	True	epithelial	male

adata[(adata.obs.cell_ontology_class == 'neutrophil') & (adata.obs.method == '10X')]

View of AnnData object with n_obs × n_vars = 138 × 58870
    obs: 'organ_tissue', 'method', 'donor', 'anatomical_information', 'n_counts_UMIs', 'n_genes', 'cell_ontology_class', 'free_annotation', 'manually_annotated', 'compartment', 'gender'
    var: 'gene_symbol', 'feature_type', 'ensemblid', 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std'
    uns: '_scvi', '_training_mode', 'cell_ontology_class_colors', 'dendrogram_cell_type_tissue', 'dendrogram_computational_compartment_assignment', 'dendrogram_consensus_prediction', 'dendrogram_tissue_cell_type', 'donor_colors', 'donor_method_colors', 'hvg', 'method_colors', 'neighbors', 'organ_tissue_colors', 'sex_colors', 'tissue_colors', 'umap'
    obsm: 'X_pca', 'X_scvi', 'X_scvi_umap', 'X_umap'
    layers: 'decontXcounts', 'raw_counts'
    obsp: 'connectivities', 'distances'

adata[(adata.obs.cell_ontology_class == 'neutrophil') &\
      ((adata.obs.method == '10X') | (adata.obs.donor == 'TSP2'))]

View of AnnData object with n_obs × n_vars = 172 × 58870
    obs: 'organ_tissue', 'method', 'donor', 'anatomical_information', 'n_counts_UMIs', 'n_genes', 'cell_ontology_class', 'free_annotation', 'manually_annotated', 'compartment', 'gender'
    var: 'gene_symbol', 'feature_type', 'ensemblid', 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std'
    uns: '_scvi', '_training_mode', 'cell_ontology_class_colors', 'dendrogram_cell_type_tissue', 'dendrogram_computational_compartment_assignment', 'dendrogram_consensus_prediction', 'dendrogram_tissue_cell_type', 'donor_colors', 'donor_method_colors', 'hvg', 'method_colors', 'neighbors', 'organ_tissue_colors', 'sex_colors', 'tissue_colors', 'umap'
    obsm: 'X_pca', 'X_scvi', 'X_scvi_umap', 'X_umap'
    layers: 'decontXcounts', 'raw_counts'
    obsp: 'connectivities', 'distances'

l = ['neutrophil', 'basophil']
adata[adata.obs.cell_ontology_class.isin(l)]
# adata[~adata.obs.cell_ontology_class.isin(l)]

View of AnnData object with n_obs × n_vars = 860 × 58870
    obs: 'organ_tissue', 'method', 'donor', 'anatomical_information', 'n_counts_UMIs', 'n_genes', 'cell_ontology_class', 'free_annotation', 'manually_annotated', 'compartment', 'gender'
    var: 'gene_symbol', 'feature_type', 'ensemblid', 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std'
    uns: '_scvi', '_training_mode', 'cell_ontology_class_colors', 'dendrogram_cell_type_tissue', 'dendrogram_computational_compartment_assignment', 'dendrogram_consensus_prediction', 'dendrogram_tissue_cell_type', 'donor_colors', 'donor_method_colors', 'hvg', 'method_colors', 'neighbors', 'organ_tissue_colors', 'sex_colors', 'tissue_colors', 'umap'
    obsm: 'X_pca', 'X_scvi', 'X_scvi_umap', 'X_umap'
    layers: 'decontXcounts', 'raw_counts'
    obsp: 'connectivities', 'distances'

adata.obs.cell_ontology_class.str.contains('t cell')

cell_id
AAACCCAAGAACTCCT_TSP14_Lung_Distal_10X_1_1                        False
AAACGAAAGGAGCTGT_TSP14_Lung_Distal_10X_1_1                        False
AAACGAACACGCGTGT_TSP14_Lung_Distal_10X_1_1                        False
AAACGAACATGCCGCA_TSP14_Lung_Distal_10X_1_1                        False
AAAGAACGTTAAGGGC_TSP14_Lung_Distal_10X_1_1                        False
                                                                  ...  
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P1_S361    False
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P2_S362    False
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P6_S366    False
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P8_S368    False
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P9_S369    False
Name: cell_ontology_class, Length: 35682, dtype: bool

adata[adata.obs.cell_ontology_class.str.contains('t cell')]

View of AnnData object with n_obs × n_vars = 2071 × 58870
    obs: 'organ_tissue', 'method', 'donor', 'anatomical_information', 'n_counts_UMIs', 'n_genes', 'cell_ontology_class', 'free_annotation', 'manually_annotated', 'compartment', 'gender'
    var: 'gene_symbol', 'feature_type', 'ensemblid', 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std'
    uns: '_scvi', '_training_mode', 'cell_ontology_class_colors', 'dendrogram_cell_type_tissue', 'dendrogram_computational_compartment_assignment', 'dendrogram_consensus_prediction', 'dendrogram_tissue_cell_type', 'donor_colors', 'donor_method_colors', 'hvg', 'method_colors', 'neighbors', 'organ_tissue_colors', 'sex_colors', 'tissue_colors', 'umap'
    obsm: 'X_pca', 'X_scvi', 'X_scvi_umap', 'X_umap'
    layers: 'decontXcounts', 'raw_counts'
    obsp: 'connectivities', 'distances'

adata[adata.obs.cell_ontology_class.str.contains('t cell')].obs

	organ_tissue	method	donor	anatomical_information	n_counts_UMIs	n_genes	cell_ontology_class	free_annotation	manually_annotated	compartment	gender
cell_id
AAAGGGCGTCGGAAAC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	3052.0	1166	cd4-positive alpha-beta t cell	cd4-positive alpha-beta t cell	True	immune	male
AACAAGAGTGTATACC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	4181.0	1659	cd8-positive alpha-beta t cell	cd8-positive alpha-beta t cell	True	immune	male
AAGGAATGTGTGGTCC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	8710.0	2606	cd8-positive alpha-beta t cell	cd8-positive alpha-beta t cell	True	immune	male
ACACGCGGTTGCCGAC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	13060.0	3888	cd4-positive alpha-beta t cell	cd4-positive alpha-beta t cell	True	immune	male
ACAGGGATCTTCTCAA_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	3023.0	1181	cd8-positive alpha-beta t cell	cd8-positive alpha-beta t cell	True	immune	male
…	…	…	…	…	…	…	…	…	…	…	…
TSP2_LungNeuron_proximal_SS2_B113240_B134457_MixedStromalNeuron_K8_S56	Lung	smartseq2	TSP2	Neuron	1947811.0	5807	respiratory goblet cell	respiratory goblet cell	True	epithelial	female
TSP2_LungNeuron_proximal_SS2_B113240_B134457_MixedStromalNeuron_M12_S108	Lung	smartseq2	TSP2	Neuron	1368759.0	4310	respiratory goblet cell	respiratory goblet cell	True	epithelial	female
TSP2_LungNeuron_proximal_SS2_B113453_B133092_MixedStromalNeuron_A5_S125	Lung	smartseq2	TSP2	Neuron	2427549.0	4968	respiratory goblet cell	respiratory goblet cell	True	epithelial	female
TSP2_LungNeuron_proximal_SS2_B113453_B133092_MixedStromalNeuron_G16_S280	Lung	smartseq2	TSP2	Neuron	129219.0	960	cd4-positive, alpha-beta t cell	cd4-positive, alpha-beta t cell	True	immune	female
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Empty_G12_S156	Lung	smartseq2	TSP2	proxmedialdistal	499949.0	5598	respiratory goblet cell	respiratory goblet cell	True	epithelial	female

2071 rows × 11 columns

adata[(adata.obs.cell_ontology_class.str.contains('t cell')) &\
     (adata.obs.cell_ontology_class != 'respiratory goblet cell')].obs

	organ_tissue	method	donor	anatomical_information	n_counts_UMIs	n_genes	cell_ontology_class	free_annotation	manually_annotated	compartment	gender
cell_id
AAAGGGCGTCGGAAAC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	3052.0	1166	cd4-positive alpha-beta t cell	cd4-positive alpha-beta t cell	True	immune	male
AACAAGAGTGTATACC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	4181.0	1659	cd8-positive alpha-beta t cell	cd8-positive alpha-beta t cell	True	immune	male
AAGGAATGTGTGGTCC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	8710.0	2606	cd8-positive alpha-beta t cell	cd8-positive alpha-beta t cell	True	immune	male
ACACGCGGTTGCCGAC_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	13060.0	3888	cd4-positive alpha-beta t cell	cd4-positive alpha-beta t cell	True	immune	male
ACAGGGATCTTCTCAA_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	3023.0	1181	cd8-positive alpha-beta t cell	cd8-positive alpha-beta t cell	True	immune	male
…	…	…	…	…	…	…	…	…	…	…	…
TTTGTTGTCAAGCCCG_TSP2_Lung_proxmedialdistal_10X_1_2	Lung	10X	TSP2	proxmedialdistal	2892.0	1349	cd4-positive, alpha-beta t cell	cd4-positive, alpha-beta t cell	True	immune	female
TTTGTTGTCTACCACC_TSP2_Lung_proxmedialdistal_10X_1_2	Lung	10X	TSP2	proxmedialdistal	3234.0	1566	cd8-positive, alpha-beta t cell	cd8-positive, alpha-beta t cell	True	immune	female
TSP2_Lung_proxmedialdistal_SS2_B113378_B104862_Endothelial_N21_S333	Lung	smartseq2	TSP2	nan	29807.0	224	cd4-positive, alpha-beta t cell	cd4-positive, alpha-beta t cell	True	immune	female
TSP2_LungNeuron_proximal_SS2_B113240_B134457_MixedStromalNeuron_C22_S178	Lung	smartseq2	TSP2	Neuron	94079.0	1065	cd4-positive, alpha-beta t cell	cd4-positive, alpha-beta t cell	True	immune	female
TSP2_LungNeuron_proximal_SS2_B113453_B133092_MixedStromalNeuron_G16_S280	Lung	smartseq2	TSP2	Neuron	129219.0	960	cd4-positive, alpha-beta t cell	cd4-positive, alpha-beta t cell	True	immune	female

1317 rows × 11 columns

Advanced cell filtering and subsetting

def fancy_filter(x):
    if x/5 > 1000:
        return True
    else:
        return False

adata.obs.n_genes.map(fancy_filter)

cell_id
AAACCCAAGAACTCCT_TSP14_Lung_Distal_10X_1_1                         True
AAACGAAAGGAGCTGT_TSP14_Lung_Distal_10X_1_1                         True
AAACGAACACGCGTGT_TSP14_Lung_Distal_10X_1_1                        False
AAACGAACATGCCGCA_TSP14_Lung_Distal_10X_1_1                        False
AAAGAACGTTAAGGGC_TSP14_Lung_Distal_10X_1_1                         True
                                                                  ...  
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P1_S361    False
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P2_S362    False
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P6_S366    False
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P8_S368    False
TSP2_Lung_proxmedialdistal_SS2_B113373_B134458_Stromal_P9_S369     True
Name: n_genes, Length: 35682, dtype: bool

adata[adata.obs.n_genes.map(fancy_filter)].obs.head(1)

	organ_tissue	method	donor	anatomical_information	n_counts_UMIs	n_genes	cell_ontology_class	free_annotation	manually_annotated	compartment	gender
cell_id
AAACCCAAGAACTCCT_TSP14_Lung_Distal_10X_1_1	Lung	10X	TSP14	Distal	44221.0	6970	type ii pneumocyte	type ii pneumocyte	True	epithelial	male

def very_fancy_filter(x):
    donor, n_counts_UMIs, n_genes = x
    if donor.endswith('14') and n_counts_UMIs/n_genes > 10:
        return True
    else:
        return False

adata[adata.obs[['donor', 'n_counts_UMIs', 'n_genes']].apply(very_fancy_filter, axis = 1)]

View of AnnData object with n_obs × n_vars = 32 × 58870
    obs: 'organ_tissue', 'method', 'donor', 'anatomical_information', 'n_counts_UMIs', 'n_genes', 'cell_ontology_class', 'free_annotation', 'manually_annotated', 'compartment', 'gender'
    var: 'gene_symbol', 'feature_type', 'ensemblid', 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std'
    uns: '_scvi', '_training_mode', 'cell_ontology_class_colors', 'dendrogram_cell_type_tissue', 'dendrogram_computational_compartment_assignment', 'dendrogram_consensus_prediction', 'dendrogram_tissue_cell_type', 'donor_colors', 'donor_method_colors', 'hvg', 'method_colors', 'neighbors', 'organ_tissue_colors', 'sex_colors', 'tissue_colors', 'umap'
    obsm: 'X_pca', 'X_scvi', 'X_scvi_umap', 'X_umap'
    layers: 'decontXcounts', 'raw_counts'
    obsp: 'connectivities', 'distances'

Gene-based cell filtering and subsetting

import numpy as np
adata.var_names == 'CDKN2A'

array([False, False, False, ..., False, False, False])

np.where(adata.var_names == 'CDKN2A')

(array([26368], dtype=int64),)

gene_loc = np.where(adata.var_names == 'CDKN2A')[0][0]
gene_loc

26368

adata.X

<35682x58870 sparse matrix of type '<class 'numpy.float32'>'
    with 119787345 stored elements in Compressed Sparse Row format>

adata.X[:, gene_loc]

<35682x1 sparse matrix of type '<class 'numpy.float32'>'
    with 1625 stored elements in Compressed Sparse Row format>

adata.X[:, gene_loc].toarray()

array([[0.],
       [0.],
       [0.],
       ...,
       [0.],
       [0.],
       [0.]], dtype=float32)

adata[adata.X[:, gene_loc].toarray() > 0]

View of AnnData object with n_obs × n_vars = 1563 × 58870
    obs: 'organ_tissue', 'method', 'donor', 'anatomical_information', 'n_counts_UMIs', 'n_genes', 'cell_ontology_class', 'free_annotation', 'manually_annotated', 'compartment', 'gender'
    var: 'gene_symbol', 'feature_type', 'ensemblid', 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std'
    uns: '_scvi', '_training_mode', 'cell_ontology_class_colors', 'dendrogram_cell_type_tissue', 'dendrogram_computational_compartment_assignment', 'dendrogram_consensus_prediction', 'dendrogram_tissue_cell_type', 'donor_colors', 'donor_method_colors', 'hvg', 'method_colors', 'neighbors', 'organ_tissue_colors', 'sex_colors', 'tissue_colors', 'umap'
    obsm: 'X_pca', 'X_scvi', 'X_scvi_umap', 'X_umap'
    layers: 'decontXcounts', 'raw_counts'
    obsp: 'connectivities', 'distances'