Chinese Immune Cell Atlas: Landmark Study Reveals Surprising Population-Specific Variations in Immune Functions (Science, 2026)

The Dawn of the Chinese Immune Multi-Omics Atlas: A Game-Changer in Immunology

In a groundbreaking achievement for Chinese biomedical research, scientists have unveiled the Chinese Immune Multi-Omics Atlas (CIMA), a comprehensive map of over 10 million immune cells from 428 healthy adults. Published in Science on January 8, 2026, this landmark study highlights profound population-specific variations in immune functions, challenging the Eurocentric bias in existing datasets. Led by teams from BGI-Research's State Key Laboratory of Genome and Multi-omics Technologies, Shanghai Jiao Tong University School of Medicine, and Shanxi Medical University, CIMA integrates single-cell RNA sequencing (scRNA-seq), single-cell ATAC sequencing (scATAC-seq), whole-genome sequencing (WGS), and plasma metabolomics to decode how genetics, age, and sex shape immunity in the Chinese population.

This atlas not only catalogs 73 distinct immune cell types—including rare subsets under 0.1% frequency—but also reveals enhancer-driven gene regulatory networks (GRNs) linking 84,625 regulatory regions to 13,645 target genes across 61 subtypes. Such depth positions Chinese universities at the forefront of precision immunology, fostering opportunities for researchers in genomics and immunotherapy.

Unprecedented Scale: Methods Behind CIMA's Massive Dataset

The CIMA project profiled peripheral blood mononuclear cells (PBMCs)—key players in adaptive and innate immunity—from participants aged 20 to 77, ensuring a broad representation of adult Chinese demographics. After rigorous quality control, 6.5 million cells from scRNA-seq and 3.8 million from scATAC-seq were analyzed, identifying 338,036 candidate cis-regulatory elements (cCREs). Techniques like iterative clustering and hierarchical annotation enabled precise cell typing, while WGS facilitated quantitative trait loci (QTL) mapping at single-cell resolution.

This multi-omics approach—combining transcriptomics, epigenomics, genomics, and biochemical profiles—overcomes limitations of prior atlases, providing a robust foundation for studying immune heterogeneity. For aspiring immunologists in China, mastering these technologies opens doors at institutions like BGI and SJTU research labs.

Population-Specific Variations: Chinese Immunity Differs from European and Japanese Counterparts

A striking revelation is the divergence in immune profiles between Chinese, European, and Japanese cohorts. While core pathways remain conserved, CIMA uncovers significant differences in cell proportions, gene expression, and chromatin accessibility—factors explaining variable immunotherapy responses across ethnicities. For instance, dynamic expression QTLs (eQTLs) in monocytes and B cells vary uniquely, linked to age and sex.

• Chinese adults show distinct regulatory T cell (Treg) dynamics influencing inflammation.
• Compared to Europeans, higher variability in natural killer (NK) cell functions.
• Japanese datasets reveal subtler shifts, underscoring East Asian diversity.

These insights, from Shanghai Jiao Tong University's Ruijin Hospital collaborators, emphasize the need for ancestry-tailored therapies.

Gene Regulatory Networks: 237 Robust Regulons Unveiled

CIMA constructed enhancer-driven GRNs, pinpointing 237 regulons governed by key transcription factors (TFs). Age-associated networks highlight declining immune vigor, while sex-specific patterns affect cytokine production. This cell-type resolved mapping—84,625 regions to 13,645 genes—illuminates how noncoding variants drive immune diversity.

Researchers at Shanxi Medical University contributed vital plasma data, integrating lipids and metabolites for holistic views. Such interdisciplinary work exemplifies China's rising prowess in systems immunology.

QTL Mapping and Pleiotropic Disease Links

Cell-resolved QTL analysis identified 9,600 eGenes and 52,361 chromatin accessibility peaks (caPeaks), with 30% eGenes and 55% caPeaks unique to one cell type. Summary-data-based Mendelian randomization (SMR) yielded 1,196 associations across 68 cell types and 154 traits, including a variant (rs34415530) tying IKZF4 in CD4+ FOXP3+ Tregs to IL-12B levels and asthma risk.

These pleiotropic links bridge genetics to diseases like autoimmune disorders, vital for Chinese clinicians.

CIMA-CLM: Revolutionary AI for Immune Prediction

The cell language model CIMA-CLM integrates chromatin sequences and scRNA-seq to predict accessibility (Pearson r=0.8951, AUROC=0.9560 across 32 types). It simulates variant effects via in silico mutagenesis, accelerating noncoding variant interpretation. BGI's AI prowess here signals booming demand for computational immunologists in Shenzhen hubs.

Implications for Precision Medicine and Immunotherapy in China

CIMA refines GWAS interpretations for immune diseases, revealing why CAR-T or checkpoint inhibitors falter in non-Europeans. For prevalent Chinese conditions like hepatitis or nasopharyngeal cancer, tailored therapies loom. Experts like Academician Guang Ning note synergies with metabolic studies at SJTU.

Open via CIMA Portal, it empowers global researchers while bolstering China's biopharma ecosystem.

Chinese Universities Driving the Frontier

Shanghai Jiao Tong University (SJTU) and Shanxi Medical University anchor CIMA, with UCAS training next-gen talent. BGI-Research, tied to Shenzhen's university network, exemplifies public-private synergy. Faculty positions in immunology surge, as seen at China's top research hubs.

Phase II and Future Horizons

Phase II targets disease cohorts (autoimmunity, CVD, infections) using Stereo-cell tech for 'virtual cells'. This positions China as immunology leader, with postdoc openings in multi-omics.

Career Opportunities in China's Immunology Boom

CIMA sparks demand for postdocs and faculty in genomics/immunology at SJTU, BGI, and beyond. Explore research jobs, postdoc roles, or career advice to join this revolution. Check professor ratings for mentors.

Photo by Олександр К on Unsplash