Billion CellxCell Project
Thank you for joining us for the 2025 Cell-Cell Symposium — a day packed with groundbreaking science, bold ideas, and exciting directions for the future of single-cell and multicell analysis. We welcomed approximately 150 attendees, including academic researchers, industry professionals, and trainees from across California and as far as Washington and Georgia.
Key Insights from the Cell-Cell Communication Symposium
If you missed it (or want a refresher), we’ve distilled the top takeaways below:
-
Nathan Lewis (UCSD) highlighted how tensor decomposition and cortical organoid models reveal causal genetic pathways in neurodevelopment and disease.
-
Rong Lu (USC) explored how hematopoietic stem and progenitor cells (HSPCs) interact with their neighbors in the bone marrow niche using spatially resolved profiling.
Introduced a signaling score derived from cell-cell proximity and secreted factor profiles to quantify interactions. -
Xiaojing Gao (Stanford) shared tools like RADARs, LIDAR, RELEASE, and CellREADR for designing synthetic circuits that sense RNA and control secretion or editing
-
Jessica Li (UCLA) presented statistical strategies for distinguishing emergent cell-cell interactions using Nanovials and scRNA-seq.
-
Qing Nie (UC Irvine) shared the evolution from CellChat to newer tools like NeuronChat, ExFINDER, and SigFlow, which infer not just who is primed to talk to whom, but how signals are predicted to flow hierarchically across tissues and respond to perturbations.
-
Matt Thomson (Caltech) introduced Morpheus and Cell Interaction Foundation Models, machine learning tools trained on spatial genomics data to predict how microenvironments shape T cell infiltration and therapy response.
Symposium Themes: Where Cell-Cell Communication Is Headed
1. Inference Isn’t Enough — We Need Validation. From scRNA-seq to spatial models, most cell-cell communication tools rely on assumptions. Researchers emphasized the need for functional validation platforms—to prove that cells truly communicate, not just co-express ligands and receptors.
2. Spatial Context Matters. Whether modeling bone marrow niches or tumor microenvironments, where cells are and who they're near deeply impacts their behavior. Tools that ignore spatial proximity risk missing the real story.
3. Temporal Flow and Signal Cascades. It’s not just about who talks to whom—it’s how signals flow over time, through layers of cellular hierarchies. Several speakers emphasized reconstructing sequential, causal pathways rather than static links.
4. Computational Models Need Experimental Testbeds. Predictive models like Morpheus and foundation GNNs are powerful—but platforms like Nanovials are needed to generate training data, test and refine their predictions in vitro at scale.
5. Synthetic Biology Is Expanding the Frontier. Tools like RADAR, LIDAR, and CellREADR are moving us from observation to manipulation—rewiring how cells send and receive information, and in so doing, understand it better.
Don’t Just Predict Communication. Capture It.
Use Partillion's Nanovial Multicell Assays to validate cell-cell communication with functional, single-cell resolution.