Cell-cell communication (CCC)¶

Introduction and installation¶

  • Tool: CellphoneDB, liana, decoupler
  • Github: None
  • Paper: https://www.sc-best-practices.org/mechanisms/cell_cell_communication.html
In [ ]:
# python libs
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

import scanpy as sc
import liana as li
import decoupler as dc

import session_info

# import cellphonedb method via liana
from liana.method import cellphonedb
from liana.method import rank_aggregate

# Set the system path
import os
import sys
sys.path.append(os.path.abspath('../../'))

from Code.Utils.CCC import *

import warnings
warnings.filterwarnings('ignore')
In [ ]:
# figure settings
sc.settings.set_figure_params(dpi=200, frameon=False)
sc.set_figure_params(dpi=200, facecolor="white")
sc.set_figure_params(figsize=(9, 5))

Load the data¶

In [ ]:
# Read in
adata = sc.read("../../data/Preprocessed_data/harmony_cellmarkers.h5ad")
adata.obs_names_make_unique()
adata
Out[ ]:
AnnData object with n_obs × n_vars = 7391 × 33694
    obs: 'batch', 'n_genes_by_counts', 'total_counts', 'total_counts_mt', 'pct_counts_mt', 'leiden', 'cell_type'
    var: 'mean', 'std', 'highly_variable', 'means', 'dispersions', 'dispersions_norm'
    uns: 'batch_colors', 'cell_type_colors', 'dendrogram_cell_type', 'hvg', 'leiden', 'leiden_colors', 'log1p', 'neighbors', 'pca', 'umap'
    obsm: 'X_pca', 'X_pca_harmony', 'X_umap', 'ora_estimate', 'ora_pvals'
    varm: 'PCs'
    layers: 'log_norm', 'norm', 'raw', 'scale_data'
    obsp: 'connectivities', 'distances'

Infer cell-cell communication¶

In [ ]:
cellphonedb(
    adata, groupby="cell_type", use_raw=False, return_all_lrs=True, verbose=True
)

Using `.X`!
Converting to sparse csr matrix!
12390 features of mat are empty, they will be removed.
['RP11-442N24__B.1', 'RP11-99J16__A.2', 'RP11-544L8__B.4', 'RP11-524D16__A.3', 'RP11-1157N2__B.2', 'RP4-633O19__A.1'] contain `_`. Consider replacing those!
Using resource `consensus`.
0.32 of entities in the resource are missing from the data.

Generating ligand-receptor stats for 7391 samples and 1150 features

100%|██████████| 1000/1000 [00:20<00:00, 49.37it/s]
In [ ]:
adata.uns["liana_res"].head()
Out[ ]:
ligand ligand_complex ligand_means ligand_props receptor receptor_complex receptor_means receptor_props source target lrs_to_keep lr_means cellphone_pvals
36440 CALM1 CALM1 0.382101 1.0 HMMR HMMR 1.675231 1.0 Memory T cell Cartilage progenitor cell True 1.028666 0.0
24210 CALM1 CALM1 0.252696 1.0 HMMR HMMR 1.675231 1.0 Cartilage progenitor cell Cartilage progenitor cell True 0.963964 0.0
31548 CALM1 CALM1 0.210917 1.0 HMMR HMMR 1.675231 1.0 Interstitial progenitor cell Cartilage progenitor cell True 0.943074 0.0
133941 TIMP1 TIMP1 0.514081 1.0 CD63 CD63 1.293064 1.0 Memory T cell Memory T cell True 0.903573 0.0
140617 PTN PTN 0.518862 1.0 PTPRS PTPRS 1.284148 1.0 Cartilage progenitor cell Regulatory chondrocyte True 0.901505 0.0
In [ ]:
li.pl.dotplot(
    adata=adata,
    colour="lr_means",
    size="cellphone_pvals",
    inverse_size=True,  # we inverse sign since we want small p-values to have large sizes
    # We choose only the cell types which we wish to plot
    source_labels=adata.uns["liana_res"].source.unique()[:3],
    target_labels=adata.uns["liana_res"].target.unique()[3:],
    # since cpdbv2 suggests using a filter to FPs
    # we can filter the interactions according to p-values <= 0.01
    filter_fun=filter_func_cellphone_pvals,
    # as this type of methods tends to result in large numbers
    # of predictions, we can also further order according to
    # expression magnitude
    orderby="lr_means",
    orderby_ascending=False,  # we want to prioritize those with highest expression
    top_n=20,  # and we want to keep only the top 20 interactions
    figure_size=(12, 6),
    size_range=(1, 6),
)

Fontsize 0.00 < 1.0 pt not allowed by FreeType. Setting fontsize = 1 pt
No description has been provided for this image
In [ ]:
li.method.show_methods()
Out[ ]:
Method Name Magnitude Score Specificity Score Reference
0 CellPhoneDB lr_means cellphone_pvals Efremova, M., Vento-Tormo, M., Teichmann, S.A....
0 Connectome expr_prod scaled_weight Raredon, M.S.B., Yang, J., Garritano, J., Wang...
0 log2FC None lr_logfc Dimitrov, D., Türei, D., Garrido-Rodriguez, M....
0 NATMI expr_prod spec_weight Hou, R., Denisenko, E., Ong, H.T., Ramilowski,...
0 SingleCellSignalR lrscore None Cabello-Aguilar, S., Alame, M., Kon-Sun-Tack, ...
0 Rank_Aggregate magnitude_rank specificity_rank Dimitrov, D., Türei, D., Garrido-Rodriguez, M....
0 Geometric Mean lr_gmeans gmean_pvals CellPhoneDBv2's permutation approach applied t...
0 scSeqComm inter_score None Baruzzo, G., Cesaro, G., Di Camillo, B. 2022. ...
0 CellChat lr_probs cellchat_pvals Jin, S., Guerrero-Juarez, C.F., Zhang, L., Cha...
In [ ]:
rank_aggregate(
    adata, groupby="cell_type", return_all_lrs=True, use_raw=False, verbose=True
)

Using `.X`!
Converting to sparse csr matrix!
12390 features of mat are empty, they will be removed.
['RP11-442N24__B.1', 'RP11-99J16__A.2', 'RP11-544L8__B.4', 'RP11-524D16__A.3', 'RP11-1157N2__B.2', 'RP4-633O19__A.1'] contain `_`. Consider replacing those!
Using resource `consensus`.
0.32 of entities in the resource are missing from the data.

Generating ligand-receptor stats for 7391 samples and 1150 features
Assuming that counts were `natural` log-normalized!
Running CellPhoneDB

100%|██████████| 1000/1000 [00:19<00:00, 51.37it/s]

Running Connectome
Running log2FC
Running NATMI
Running SingleCellSignalR
In [ ]:
adata.uns["liana_res"].drop_duplicates(
    ["ligand_complex", "receptor_complex"]
).head()
Out[ ]:
source target ligand_complex receptor_complex lr_means cellphone_pvals expr_prod scaled_weight lr_logfc spec_weight lrscore specificity_rank magnitude_rank
0 CD4+ T cell CD4+ T cell ACE BDKRB2 -0.023789 0.686 0.000550 -0.018248 -4.152555 0.024103 NaN 1.000000 NaN
1 CD4+ T cell CD4+ T cell ACTR2 ADRB2 -0.025335 0.854 -0.000715 -0.025335 -0.229537 -0.072139 NaN 0.983755 NaN
2 CD4+ T cell CD4+ T cell ACTR2 LDLR -0.023855 0.842 -0.000681 -0.023855 -0.017867 -0.135331 NaN 1.000000 NaN
3 CD4+ T cell CD4+ T cell ADA DPP4 0.013636 0.083 -0.001886 0.014201 -2.353680 -0.080512 NaN 1.000000 NaN
4 CD4+ T cell CD4+ T cell ADAM10 AXL -0.000811 0.509 -0.013606 -0.000811 -0.052569 0.042280 NaN 0.268218 NaN
In [ ]:
li.pl.dotplot(
    adata=adata,
    colour="magnitude_rank",
    size="specificity_rank",
    inverse_colour=True,  # we inverse sign since we want small p-values to have large sizes
    inverse_size=True,
    # We choose only the cell types which we wish to plot
    source_labels=['CD4+ T cell'],
    target_labels=["CD4+ T cell"],
    # since the rank_aggregate can also be interpreted as a probability distribution
    # we can again filter them according to their specificity significance
    # yet here the interactions are filtered according to
    # how consistently highly-ranked is their specificity across the methods
    filter_fun = filter_func_specificity_rank,
    # again, we can also further order according to magnitude
    orderby="magnitude_rank",
    orderby_ascending=True,  # prioritize those with lowest values
    top_n=20,  # and we want to keep only the top 20 interactions
    figure_size=(9, 5),
    size_range=(1, 6),
)

Fontsize 0.00 < 1.0 pt not allowed by FreeType. Setting fontsize = 1 pt
No description has been provided for this image