CellPhoneDB is a publicly available repository of HUMAN curated receptors, ligands and their interactions paired with a tool to interrogate your own single-cell transcriptomics data (or even bulk transcriptomics data if your samples represent pure populations!).

A distictive feature of CellPhoneDB is that the subunit architecture of either ligands and receptors is taken into account, representing heteromeric complexes accurately. This is crucial, as cell communication relies on multi-subunit protein complexes that go beyond the binary representation used in most databases and studies. CellPhoneDB also incorporates biosynthetic pathways in which we use the last representative enzyme as a proxy of ligand abundance, by doing so, we include interactions involving non-peptidic CellPhoneDB includes only manually curated & reviewd molecular interactions with evidenced role in cellular communication.

For more details on using CellPhoneDB for scRNA-seq data analysis, check the DOCUMENTATION.

1. New python package that can be easily executed in Jupyter Notebook and Collabs.
2. A new method to ease the query of CellPhoneDB results.
3. Tutorials to run CellPhoneDB (available here)
4. Improved computational efficiency of method 2 cpdb_statistical_analysis_method.
5. A new database (cellphonedb-data v4.1.0) with more manually curated interactions, making up to a total of ~3,000 interactions. This release of CellphoneDB database has three main changes:
   • Integrates new manually reviewed interactions with evidenced roles in cell-cell communication.
   • Includes non-protein molecules acting as ligands.
   • CellphoneDB does not longer imports interactions from external resources. This is to avoid the inclusion of low-confidence interactions.

See updates from previous releases here.

We highly recommend using an isolated python environment (as described in steps 1 and 2) using conda or virtualenv but you could of course omit these steps and install via pip immediately.

1. Create python=>3.8 environment

   • Using conda: conda create -n cpdb python=3.8
   • Using virtualenv: python -m venv cpdb

2. Activate environment

   • Using conda: source activate cpdb
   • Using virtualenv: source cpdb/bin/activate

3. Install CellPhoneDB pip install cellphonedb

4. Set up the kernel for the Jupyter notebooks.

   • Install the ipython kernel: pip install -U ipykernel.
   • Add the environment as a jupyter kernel: python -m ipykernel install --user --name 'cpdb'.
   • Open/Start Jupyter and select the created kernel.

5. Download the database.

   • Follow this tutorial.

NOTE: Works with Python v3.8 or greater. If your default Python interpreter is for v2.x (you can check it with python --version), calls to python/pip should be substituted by python3/pip3.

Please, activate your environment if you didn't previously

• Using conda: source activate cpdb
• Using virtualenv: source cpdb/bin/activate

We have created a set of tutorials that can be accessed for each To use the example data, please tutorials and data.

Counts file can be a text file or a h5ad (recommended), h5 or a path to a folder containing a 10x output with mtx/barcode/features files. NOTE: Your gene/protein ids must be HUMAN. If you are working with another specie such as mouse, we recommend you to convert the gene ids to their corresponding orthologous.

This is a two columns file indicanting which gene is specific or upregulated in a cell type (see example ). The first column should be the cell type/cluster name (matching those in meta.txt) and the second column the associated gene id. The remaining columns are ignored. We provide notebooks for both Seurat and Scanpy users. It is on you to design a DEG analysis appropiated for your research question.

This is a two columns file indicating which cell type is in which spatial microenvironment (see example ). CellphoneDB will use this information to define possible pairs of interacting cells (i.e. pairs of clusters co-appearing in a microenvironment).

For more detailed examples refer to out tutorials here.

from cellphonedb.src.core.methods import cpdb_degs_analysis_method

deconvoluted, means, relevant_interactions, significant_means = cpdb_degs_analysis_method.call(
        cpdb_file_path = cellphonedb.zip,
        meta_file_path = test_meta.txt,
        counts_file_path = test_counts.h5ad,
        degs_file_path = degs_file_path,
        counts_data = 'hgnc_symbol',
        threshold = 0.1,
        output_path = out_path)

from cellphonedb.src.core.methods import cpdb_statistical_analysis_method

deconvoluted, means, pvalues, significant_means = cpdb_statistical_analysis_method.call(
        cpdb_file_path = cellphonedb.zip,
        meta_file_path = test_meta.txt,
        counts_file_path = test_counts.h5ad,
        counts_data = 'hgnc_symbol',
        output_path = out_path)

• Using text files

from cellphonedb.src.core.methods import cpdb_analysis_method

means, deconvoluted = cpdb_analysis_method.call(
        cpdb_file_path = cellphonedb.zip,
        meta_file_path = test_meta.txt,
        counts_file_path = test_counts.txt,
        counts_data = 'hgnc_symbol',
        output_path = out_path)

• Using h5ad count file

from cellphonedb.src.core.methods import cpdb_analysis_method

means, deconvoluted = cpdb_analysis_method.call(
        cpdb_file_path = cellphonedb.zip,
        meta_file_path = test_meta.txt,
        counts_file_path = test_counts.h5ad,
        counts_data = 'hgnc_symbol',
        output_path = out_path)

from cellphonedb.src.core.methods import cpdb_analysis_method

means, deconvoluted = cpdb_analysis_method.call(
        cpdb_file_path = cellphonedb.zip,
        meta_file_path = test_meta.txt,
        counts_file_path = test_counts.h5ad,
        counts_data = 'hgnc_symbol',
        microenvs_file_path = microenvs_file_path,
        output_path = out_path)

from cellphonedb.src.core.methods import cpdb_degs_analysis_method

deconvoluted, means, relevant_interactions, significant_means = cpdb_degs_analysis_method.call(
        cpdb_file_path = cellphonedb.zip,
        meta_file_path = test_meta.txt,
        counts_file_path = test_counts.h5ad,
        counts_data = 'hgnc_symbol',
        microenvs_file_path = microenvs_file_path,
        output_path = out_path)

To understand the different analysis and results, please check the results documentation.

~ Optional Method parameters:

• counts_data: [ensembl | gene_name | hgnc_symbol] Type of gene identifiers in the counts data
• iterations: Number of iterations for the statistical analysis [1000]
• threshold: % of cells expressing the specific ligand/receptor
• result_precision: Number of decimal digits in results [3]
• output_path: Directory where the results will be allocated (the directory must exist) [out]
• output_suffix: Output format of the results files (time stamp will be added to filename if not present) [txt]
• subsampling: Enable subsampling
• subsampling_log: Enable subsampling log1p for non log-transformed data inputs !!mandatory!!
• subsampling_num_pc: Subsampling NumPC argument (number of PCs to use) [100]
• subsampling_num_cells: Number of cells to subsample the dataset [1/3 of cells]

~ Optional Method Statistical parameters

• microenvs_file_path: Spatial microenviroments input file. Restricts the cluster/cell_type interacting pairs to the cluster/cell_type sharing a microenviroment (i.e. only test a combination of clusters if these coexist in a microenviroment). This file should contain two columns: 1st column indicates the cluster/cell_type, 2nd column indicates the microenviroment name. See example here.
• pvalue: P-value threshold [0.05]
• debug_seed: Debug random seed -1. To disable it please use a value >=0 [-1]
• threads: Number of threads to use. >=1 [4]

CellPhoneDB results can be queried by making use of the search_analysis_results method. This method requires two of the files generated by CellPhoneDB; significant_means and deconvoluted.

Through this method, users can specify the cell pairs of interest and both; the genes query_genes participating in the interaction and/or the name of the interaction itself query_interactions. This method will search for significant/relevant interactions in which any cell specified in query_cell_types_1 is found to any cell specified in query_cell_types_2. Cell pairs within any of these two lists will not be queried, that is to say, no interaction between cells A and B or C and D will be queried.

from cellphonedb.utils import search_utils

search_results = search_utils.search_analysis_results(
    query_cell_types_1 = list_of_cells_1,
    query_cell_types_2 = list_of_cells_2,
    query_genes = list_of_genes,
    query_interactions = list_of_interaction_names,
    significant_means = significant_means,
    deconvoluted = cpdb_deconvoluted,
    separator = '|',
    long_format = True
)

Examples of this are provided in the [tutorials](notebooks).

Currently CellPhoneDB relies on external plotting implementations to represent the results. Some examples are provided in the tutorials.

Currently we recommend using tools such as: seaborn, ggplot or a more specific and tailored implementation as the ktplots: @zktuong:

• ktplots (R; preferred)
• ktplotspy (python; under development)

CellPhoneDB databases can be updated from the remote repository through our tool. Furthermore, available versions can be listed and downloaded for use. Please, refer to our tutorials for a comprehensive example.

First, the user must download the database to its preferred directory, once this is done, the user must provide the argument cpdb_file_path to the CellPhoneDB method to be executed with the provided version of the database.

The database is downloaded in a zip format along with the input files employed to generate it. These input files can be modified to update the database with new interactions.

The command to list available versions from the remote repository is:

from IPython.display import HTML, display
from cellphonedb.utils import db_releases_utils

display(HTML(db_releases_utils.get_remote_database_versions_html()['db_releases_html_table']))

See examples.

The command to download a version from the remote repository is:

from cellphonedb.utils import db_utils

db_utils.download_database(cpdb_target_dir, cpdb_version)

See examples.

A user can generate custom databases and use them. In order to generate a new database, a user can provide his/her own lists.

We recommend first to download CellPhoneDB database, move these downloaded files into a new folder and then modify its content to add new interactions. Once this process is completed, the created_db method will create a new database in zip format in the same folder where the inputs are located. Examples of how to download and create the database can be found here example.

To generate such a database the user has to issue this command:

from cellphonedb.utils import db_utils

db_utils.create_db(cpdb_input_dir) 

Result database file is generated in the folder out with cellphonedb_{datetime}.zip. Do not change the name of the input files, otherwise CellPhoneDB will not recognize them and and error will be thrown.

CellPhoneDB is an open-source project. If you are interested in contributing to this project, please let us know.

You can check all project documentation in the docs section

The first version of CellphoneDB was originally developed at the Teichmann Lab in the Wellcome Sanger Institute (Cambridge, UK) by Roser Vento-Tormo and Mirjana Efremova. Currently, it is being further developed and supported by the Vento-Tormo Lab (CellphoneDB ≥v3).

If you use CellphoneDB or CellphoneDB-data, please cite our papers:

• CellPhoneDB v1 (original): Single-cell reconstruction of the early maternal-fetal interface in humans. Vento-Tormo R, Efremova M, et al., Nature. 2018 link

• CellPhoneDB v2: Inferring cell-cell communication from combined expression of multi-subunit receptor-ligand complexes. Efremova M, Vento-Tormo M, Teichmann S, Vento-Tormo R. Nat Protoc. 2020 link

• CellphoneDB v3: Mapping the temporal and spatial dynamics of the human endometrium in vivo and in vitro. L Garcia-Alonso, L-François Handfield, K Roberts, K Nikolakopoulou et al. Nature Genetics 2021 link

• CellphoneDB v4 (latest): Single-cell roadmap of human gonadal development. L Garcia-Alonso, V Lorenzi et al. 2022 Nature link