Systematic mapping of emergent transcriptional states in interacting single-cell dyads by Cell-Cell-seq
Cell-cell interactions drive rapid and heterogeneous changes in gene expression, yet most transcriptomic methods either dissociate cells, losing pair identity and interaction timing, or infer communication indirectly from ligand-receptor co-expression. Here we present Cell-Cell-seq, a scalable workflow for profiling defined cell pairs ("dyads") with single-cell transcriptomic resolution. Cell-Cell-seq uses cavity-containing hydrogel microparticles (Nanovials) to confine two cells, synchronize contact onset, protect fragile conjugates during handling and sorting, and interface directly with droplet-based RNA sequencing. Using antigen-matched prostate tumor cells and engineered T cells as a model system, Cell-Cell-seq captured thousands of tumor-T cell dyads and revealed broad functional and transcriptional heterogeneity across interactions. Dyads unmasked transient activation programs that were obscured in standard well-plate co-culture, consistent with asynchronous contact in bulk assays. To distinguish interaction-induced programs from the composite nature of dyad transcriptomes, we developed a pseudo-mixing framework that generates in silico pseudo-dyads to construct an empirical null distribution under "no interaction," enabling statistically robust identification of emergent genes and partner-resolved attribution of responses. Dyad-resolved analysis further revealed coordinated cross-cell programs, including coupled chemokine expression consistent with bidirectional paracrine signaling and inverse coupling between tumor immunoregulatory programs and T cell activation. Finally, we introduce ccRepair to correct compositional dilution in mixed transcriptomes, improving interpretability while preserving genuine cross-cell coordination. Together, Cell-Cell-seq provides a generalizable platform for dissecting immune synapse biology and mapping interaction-dependent programs across heterogeneous cell populations, with applications in profiling tumor-immune communication and functionally screening immunotherapies.