← Back to Home

Which Cells Carry Disease Risk? More Than 5 Million Immune Cells Map Genetic Regulation

TenK10K pairs whole-genome and single-cell data from 1,925 people to investigate how common and rare variants alter gene expression. The atlas locates disease-associated genetic signals more precisely within specific cell types and offers leads for therapeutic target research that remain to be validated.

By SURL BioNews

Finding a DNA variant associated with disease is often just the beginning: which gene does it affect, and in which type of cell does it act? The first phase of the TenK10K study, published in Nature on October 28, compared gene expression in more than 5 million immune cells with individual genomes, seeking to bridge the gap between disease risk and cellular function.

The study included whole-genome sequencing from 1,925 people and single-cell RNA sequencing data covering 28 types of immune cells. Two records in the European Genome-phenome Archive (EGA) confirm that these datasets came from matched participants. Whole-genome sequencing achieved an average coverage depth of 30-fold, providing a basis for identifying common and rare variants. The RNA data allowed the team to observe how gene expression levels in different cell types change with genetic differences.

This type of analysis is called expression quantitative trait locus analysis, or eQTL analysis. It searches for statistical associations between DNA variants and gene expression levels. The team used the SAIGE-QTL method to model single-cell data directly, aiming to capture differences that can be obscured in signals from mixed cell populations, particularly low-frequency and rare variants that have previously been more difficult to study.

According to information from the Garvan Institute of Medical Research published through News-Medical, the study identified more than 180,000 effects of common and rare variants on gene expression, spanning approximately 21,000 genes. Among these, nearly 5,000 genes showing signals of genetic regulation had not been identified in the largest previous whole-blood study. These counts refer to regulatory effects and cannot be interpreted directly as the discovery of an equivalent number of disease-causing mutations.

Distinguishing cell types is crucial because approximately half of the regulatory effects occur in only one or a few cell types. When different cells are examined together, localized changes may be diluted by averaging. This also illustrates why the significance of a given genetic difference needs to be understood in the context of cell type and state to better approximate its actual role in the human body.

For drug research, the atlas can help narrow the search by connecting disease-associated genetic regions to potentially affected genes and cells. However, statistical associations with expression levels are insufficient to establish disease mechanisms, much less guarantee that regulating a particular gene can safely treat disease. Further validation through functional experiments and clinical studies is still required.

This resource also has limits to its applicability. The participants analyzed in the paper were of European ancestry, and the samples came from peripheral blood mononuclear cells, so its explanatory power cannot be directly extrapolated to other populations and tissues. Garvan's account states that the team plans to expand participant numbers, the range of diseases covered, and genetic diversity. These expansions will help test whether the regulatory relationships mapped so far extend to the disease experiences of more people.

References

  1. Nature
  2. News-Medical (Garvan Institute of Medical Research release)
  3. European Genome-phenome Archive
  4. European Genome-phenome Archive