MR-CCC: Bayesian Mendelian Randomization for Causal Cell--Cell Communication

2026-04-26

AI Summary

Existing ligand–receptor co-expression–based methods for inferring intercellular communication struggle to distinguish causal signaling from confounding factors. To address this, this study proposes the MR-CCC framework, which introduces receptor-abundance–modulated interaction effects for the first time. Leveraging cis-eQTLs as instrumental variables, MR-CCC integrates Bayesian Mendelian randomization, spike-and-slab priors, and efficient Gibbs sampling to jointly estimate both the main ligand effects and their interactions with receptors, thereby effectively controlling false positives. Applied to the OneK1K dataset, the method identified eight causal communication signals along the NK cell–to–monocyte axis, implicating GABA, interferon, interleukin, and prostaglandin pathways. Notably, it revealed stoichiometry-dependent dissociation of IL-18 receptor chains and cooperative action of the dual IFN-γ receptor subunits.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Cell--cell communication (CCC) is commonly inferred from ligand--receptor co-expression, an associational paradigm that cannot distinguish causal signaling from shared regulation or confounding. We propose MR-CCC, a Bayesian Mendelian randomization framework that uses cis-eQTLs as instruments for ligand and receptor expression and explicitly models receptor-modulated ligand effects through an interaction term, so the causal effect of a ligand can vary with receptor abundance. A spike--and--slab prior yields posterior inclusion probabilities quantifying evidence for causal signaling, and an efficient Gibbs sampler provides scalable inference. Benchmarked against naive regression, MVMR, and MR-BMA, MR-CCC controls false discoveries under confounding while retaining high power, and uniquely estimates both the ligand main and receptor-modulated interaction effects. Applied to the OneK1K NK cells $\to$ monocytes axis, MR-CCC identifies eight discoveries across GABA, interferon, interleukin, and prostaglandin signaling, including a stoichiometry-dependent dissociation of the two IL-18 receptor chains and co-discovery of both obligate IFN-$γ$ receptor subunits.

Problem

• cell-cell communication
• causal inference
• Mendelian randomization
• ligand-receptor
• confounding

Innovation

• Mendelian Randomization
• Bayesian inference
• cell-cell communication
• causal interaction
• eQTL