T-cells, B-cells, and natural killer cells are crucial for maintaining human health. While white blood cell counts are clinically significant, standard assays don't differentiate between these cell types. This lack of detailed information hinders our understanding of the genetic basis of immune responses and the prevalence of autoimmune diseases. Previous research suggested estimating T-cell fractions from genome sequencing data by analyzing the T-cell Receptor Alpha (TRA) locus. This study aimed to address the gap in our understanding of T-cell genetic architecture and its phenotypic consequences by estimating T-cell fractions from whole genome sequence data of over 200,000 individuals from the TOPMed and All of Us cohorts. The researchers then performed genetic association analyses and phenotypic association studies to investigate the regulation of T-cells across populations and the phenotypic consequences of natural variation in T-cell proportions. The study’s large sample size and multi-ancestry approach are crucial for capturing the complexities of human genetic variation and its effects on immune cell proportions.

Existing large-scale genome-wide association studies (GWAS) of blood cell counts haven't differentiated between T-cells, B-cells, and NK cells due to limitations in standard clinical assays. This lack of T-cell-specific data has prevented detailed analyses of T-cell genetic architecture. However, recent work by Bentham et al. demonstrated the feasibility of estimating T-cell fractions directly from genome sequencing data by assessing the depth of coverage in the V(D)J recombination region of the TRA locus. This method, which has been extensively validated previously, provided the basis for the current study. The study builds upon previous work demonstrating ancestry-specific variations in blood cell traits but extends this analysis to T-cell specific proportions for a more detailed understanding of the immune system’s genetic basis.

The study utilized whole genome sequencing (WGS) data from over 200,000 individuals in the TOPMed and All of Us cohorts. T-cell fractions were estimated using the T-cell ExTRECT tool, which quantifies T-cell fraction using a modified read depth ratio in the TRA gene, accounting for V(D)J recombination. The tool's accuracy was validated by comparing results from different exome capture kits. Genetic association analyses were conducted using linear mixed models, controlling for age, sex, and ancestry principal components. Separate analyses were performed for TOPMed and All of Us cohorts, followed by a meta-analysis to increase power. Phenotypic associations were investigated using ordinary least squares models and logistic regression, leveraging electronic health records (EHR) data. Polygenic risk scores (PRS) were generated to assess the cumulative effect of multiple genetic variants on T-cell fraction and their association with various clinical phenotypes using LabWAS. Analysis of T-cell dynamics during pregnancy was also performed using a subset of All of Us participants. The study utilized various statistical methods including linear mixed models, meta-analysis, logistic regression, ordinary least squares models, and polygenic risk score analysis.

The study identified 27 loci significantly associated with T-cell fraction. T-cell fraction showed significant correlations with sex (higher in females) and age (lower with increasing age). Analysis revealed significant associations between T-cell fraction and estimated global ancestry, with variations across different ancestral populations. The heritability of T-cell fraction was estimated to be as high as 42% in individuals of African ancestry, with chromosome 1 accounting for a substantial portion of this heritability. The study replicated several previously known genetic associations and identified novel loci associated with T-cell fraction. Analysis of EHR data revealed associations between T-cell fraction and multiple immune markers, blood markers, and a metabolic marker, consistent with the biological role of T cells. Pregnant participants showed significantly lower T-cell fractions compared to non-pregnant controls, with the lowest fraction observed at the end of the second trimester. Alterations in T-cell fraction were identified in various populations, with differing effects throughout pregnancy, suggesting a role in immune response during pregnancy.

This study is the first large-scale investigation of the genetic architecture of T-cell fraction. The findings highlight the high heritability of this trait and identify multiple genetic loci influencing T-cell abundance and diverse ancestry-specific associations. The associations between T-cell fraction and various clinical phenotypes underscore its role in health and disease. The sex and age-related differences observed are consistent with known patterns of immune system changes throughout life. The study's findings expand our understanding of the genetic and phenotypic landscape of T-cell regulation and provide valuable insights into the interplay between genetics, immune function, and health outcomes. The observed changes during pregnancy provide further insights into immune system dynamics during this period and warrant further investigation into the underlying mechanisms.

This study successfully quantified the genetic architecture of T-cell fraction at an unprecedented scale using a robust method adaptable to genome sequencing data. The identification of germline variants and phenotypic consequences associated with T-cell fraction expands our understanding of immune system regulation and its impact on human health. Future research should focus on validating these findings in larger and more diverse cohorts, exploring the role of more complex genetic variants, and investigating the underlying mechanisms connecting genetics to T-cell abundance and disease susceptibility. Expanding this approach to other immune cell types could further enhance our knowledge of immune system complexity.

While the study included a large and diverse sample, East Asian and South Asian individuals remain underrepresented. The sample skewed towards older adults, potentially limiting the generalizability of certain findings. The analysis focused on single nucleotide variants and small indels, neglecting potentially important contributions from larger variants like copy number variations and short tandem repeats. The ancestry classification method used might have introduced confounding effects in ancestry-specific analyses. While the T-cell ExTRACT method has been validated, its application to whole genome sequencing, compared to exome sequencing, still requires further comprehensive validation.