Skip to main content

LIANA+: an all-in-one cell-cell communication framework

main GitHub issues Documentation Status codecov Downloads

LIANA+ is a scalable framework that adapts and extends existing methods and knowledge to study cell-cell communication in single-cell, spatially-resolved, and multi-modal omics data. It is part of the scverse ecosystem, and relies on AnnData & MuData objects as input.

Contributions

We welcome suggestions, ideas, and contributions! Please do not hesitate to contact us, open issues, and check the contributions guide.

Vignettes

A set of extensive vignettes can be found in the LIANA+ documentation.

Claude Code Skill

LIANA+ provides an Agent Skill for Claude Code that guides the agent to use the library correctly: which method fits which data, where results land, and the defaults that trip users up. Install it once into your personal skills directory:

liana-install-skills

The command comes with pip install liana, so run it from any Python environment that has liana. It copies the skill to ~/.claude/skills/liana/, where Claude Code picks it up automatically in every project and directory; type /skills in Claude Code to confirm liana is listed. After upgrading liana, refresh the copy with liana-install-skills --force. Then ask Claude Code for liana tasks in plain words ("which cell types signal to which in this Xenium slide", "compare interactions between conditions", "run this on mouse") and the skill is consulted automatically.

Decision Tree

Use the tree below to find a starting point for your analysis. Broad, data-driven choices sit at the top and trickle down to specific methods (click a node to open its tutorial).

flowchart TD
    Start{What is your data?}

    %% ===== Spatially-resolved =====
    Start -->|Spatially-resolved| Res{Resolution?}
    Res -->|Single-cell| ScType{Analysis type?}
    ScType -->|Interaction scoring| Inflow[Inflow Score]
    ScType -->|Interaction scoring| ScConstr[Standard LR methods<br/>spatially-constrained]
    ScType -->|Spatial co-occurrence| LRIC[LRIC]
    ScType -->|Unsupervised| InflowMofa[Communication Programs<br/>Inflow + MOFA-Flex]
    Res -->|Spot-based| SpType{Analysis type?}
    SpType -->|Bivariate| LocalQ{Local<br/>interactions?}
    LocalQ -->|Yes| Local[Local Bivariate Metrics]
    LocalQ -->|No| Global[Global Bivariate Metrics]
    SpType -->|Unsupervised| MISTy[Multi-view Learning<br/>MISTy]

    %% ===== Dissociated single-cell =====
    Start -->|Dissociated single-cell| Compare{Compare across<br/>samples?}
    Compare -->|No| Steady[Steady-state LR Inference]
    Compare -->|Yes| Contrast{Specific<br/>contrast?}
    Contrast -->|Yes| Targeted[Targeted Differential]
    Contrast -->|Yes| CrossTalk[pyCrossTalkeR<br/>network differential]
    Contrast -->|No| MOFA[MOFA+]
    Contrast -->|No| Tensor[Tensor-cell2cell]
    Tensor --> TensorExt[Extended Tutorials<br/>ccc-protocols]
    Tensor --> TensorMet[CCC Patterns: protein + metabolite<br/>Tensor-cell2cell CTCA]

    %% ===== Multi-modal =====
    Start -->|Multi-modal| ModalSp{Spatial?}
    ModalSp -->|Yes| SMA[Multi-Modal Spatial]
    ModalSp -->|No| SCMulti[Multi-Modal Single-Cell]
    SCMulti --> Metab[Metabolite-mediated CCC]

    %% ===== Styling =====
    classDef decision fill:#f6f0f1,stroke:#364159,stroke-width:1px,color:#364159;
    classDef spatial fill:#e0fbf8,stroke:#00786e,color:#05564f;
    classDef dissoc fill:#fdeef4,stroke:#ba1b57,color:#7c1039;
    classDef multimodal fill:#e9ecf3,stroke:#4a587a,color:#2a3348;
    classDef external fill:#ffffff,stroke:#9e9e9e,stroke-dasharray:5 3,color:#424242;

    class Start,Res,ScType,SpType,LocalQ,Compare,Contrast,ModalSp decision;
    class Inflow,LRIC,ScConstr,InflowMofa,Local,Global,MISTy spatial;
    class Steady,Targeted,CrossTalk,MOFA,Tensor dissoc;
    class SMA,SCMulti,Metab multimodal;
    class TensorExt,TensorMet external;

    %% ===== Links (click events) =====
    click Inflow "https://liana.readthedocs.io/en/latest/tutorials/notebooks/inflow_score.html"
    click LRIC "https://liana.readthedocs.io/en/latest/tutorials/notebooks/LRIC_tutorial.html"
    click InflowMofa "https://liana.readthedocs.io/en/latest/tutorials/notebooks/inflow_mofaflex.html"
    click ScConstr "https://liana.readthedocs.io/en/latest/tutorials/notebooks/inflow_score.html"
    click Local "https://liana.readthedocs.io/en/latest/tutorials/notebooks/bivariate.html"
    click Global "https://liana.readthedocs.io/en/latest/tutorials/notebooks/bivariate.html"
    click MISTy "https://liana.readthedocs.io/en/latest/tutorials/notebooks/misty.html"
    click Steady "https://liana.readthedocs.io/en/latest/tutorials/notebooks/basic_usage.html"
    click Targeted "https://liana.readthedocs.io/en/latest/tutorials/notebooks/targeted.html"
    click CrossTalk "https://liana.readthedocs.io/en/latest/tutorials/notebooks/liana_pyCrossTalkeR.html"
    click MOFA "https://liana.readthedocs.io/en/latest/tutorials/notebooks/mofatalk.html"
    click Tensor "https://liana.readthedocs.io/en/latest/tutorials/notebooks/liana_c2c.html"
    click TensorExt "https://ccc-protocols.readthedocs.io/en/latest/"
    click TensorMet "https://earmingol.github.io/cell2cell/tutorials/Version2/Tensor-cell2cell-CTCA-LIANA/"
    click SMA "https://liana.readthedocs.io/en/latest/tutorials/notebooks/sma.html"
    click SCMulti "https://liana.readthedocs.io/en/latest/tutorials/notebooks/sc_multi.html"
    click Metab "https://liana.readthedocs.io/en/latest/tutorials/notebooks/sc_multi.html#metabolite-mediated-ccc-from-transcriptomics-data"

This tree is a guide rather than an exhaustive map: the methods are modular and can be adapted or combined across data types and questions, and all of them typically build on curated prior knowledge (see the prior knowledge tutorial for working with ligand–receptor and other resources).

API

For further information please check LIANA's API documentation.

Cite LIANA+:

@article {Dimitrov2024,
    author = {Dimitrov, Daniel and Sch{\"a}fer, Philipp Sven Lars and  Farr, Elias and Rodriguez-Mier, Pablo and Lobentanzer, Sebastian and Badia-i-Mompel, Pau and Dugourd, Aurelien and Tanevski, Jovan and Ramirez Flores, Ricardo Omar and Saez-Rodriguez, Julio},
    title = {LIANA+ provides an all-in-one framework for cell--cell communication inference},
    journal = {Nature Cell Biology},
    year = {2024},
    volume = {26},
    number = {9},
    pages = {1613--1622},
    DOI = {10.1038/s41556-024-01469-w},
    URL = {https://doi.org/10.1038/s41556-024-01469-w}
}
@article {Dimitrov2022,
    author = {Dimitrov, Daniel and T{\"u}rei, D{\'e}nes and Garrido-Rodriguez, Martin and Burmedi, Paul L. and Nagai, James S. and Boys, Charlotte and Ramirez Flores, Ricardo O. and Kim, Hyojin and Szalai, Bence and Costa, Ivan G. and Valdeolivas, Alberto and Dugourd, Aur{\'e}lien and Saez-Rodriguez, Julio},
    title = {Comparison of methods and resources for cell-cell communication inference from single-cell RNA-Seq data},
    journal = {Nature Communications},
    year = {2022},
    volume = {13},
    number = {1},
    pages = {3224},
    DOI = {10.1038/s41467-022-30755-0},
    URL = {https://doi.org/10.1038/s41467-022-30755-0}
}

Please also consider citing any of the methods and/or resources that were particularly relevant for your research!

Metadata

Release files for liana 2.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for liana 2.1.0
File Size Uploaded Provenance Actions
liana-2.1.0.tar.gz 1.6 MB With provenance Details for liana-2.1.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for liana 2.1.0
File Interpreter ABI Platform Actions
liana-2.1.0-py3-none-any.whl Python 3 none any Details for liana-2.1.0-py3-none-any.whl

Total release size: 2.3 MB

Release files / liana-2.1.0.tar.gz

Download URL liana-2.1.0.tar.gz
Size 1.6 MB
Tags Source
SHA-256 checksum
How to use checksums
84bfcafff443833dfe9f4f87492645a09433b98f18003baf53c89a2319962edd
BLAKE2b-256 checksum
How to use checksums
1048b9c5f75eb6e34a99428c109472906107d0d34436a9f937e7a344e62c9f96
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 31, 2026.

Transparency log

Release files / liana-2.1.0-py3-none-any.whl

Download URL liana-2.1.0-py3-none-any.whl
Size 664.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
4a0bd7053eba20ebeb0829b394748cd4d511e146338d92b2adf44cfc6fca9e37
BLAKE2b-256 checksum
How to use checksums
eba3bb7839937c307fb292bc54aba3235cc09cb4b0d97427febc4beefdfed0a6
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 31, 2026.

Transparency log

Security for liana 2.1.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for liana-2.1.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 31, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

2.1.0 This release

2 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page