Age-related chromosomal abnormalities in blood cells have been known for a long time, but studying them was difficult: standard methods failed to pick up changes affecting only a small fraction of cells. This work sits alongside studies that link such abnormalities to the risk of blood cancer and other diseases, as well as genetic studies of Y chromosome loss. However, for most chromosomes the genetic basis remained unknown. A new computational method made it possible to accurately measure even minor deviations in chromosome copy number from sequencing data, which opened the way to large-scale analysis.
The researchers developed a method for cleaning technical noise from whole-genome sequencing data that allows them to estimate the fraction of cells with an abnormal chromosome number to within 0.1%. This approach was applied to data from 490,089 UK Biobank participants and 350,764 All of Us participants. For each chromosome, they calculated the deviation of coverage (the number of DNA reads) from the expected value. They then performed a genome-wide association search between these deviations and millions of inherited genetic variants, and also estimated heritability using twins and siblings.
Loss of the X chromosome in women and the Y chromosome in men turned out to be highly heritable: the correlation in monozygotic twins reached 0.60-0.62, and heritability is estimated at 52-62%. For these traits, 245 and 828 genetic loci were identified, respectively. For ten types of autosomal aneuploidies (extra or missing copies of other chromosomes), 173 new loci were found. Key findings include variants in kinetochore and centromere genes that affect different chromosomes in different ways, as well as the repetitive regions ICCE and DXZ4 on the X chromosome, whose length changes the risk of X loss severalfold. Mendelian randomization did not confirm a causal link between sex chromosome loss and most diseases, apart from a weak association with prostate cancer.
The work shows for the first time that the genetic architecture of different chromosomal aneuploidies in the blood differs substantially. Unlike previous studies, which focused on individual chromosomes or used less sensitive methods, this one applied a unified approach to all chromosomes and produced quantitative estimates. This makes it possible to compare the mechanisms of instability across different chromosomes and to identify specific vulnerabilities, for example those linked to centromeric sequences. The results also cast doubt on the widespread belief that Y chromosome loss strongly affects health.
Further research could aim to uncover the molecular mechanisms behind the associations found, especially for centromeric haplotypes and repetitive regions. More accurate assemblies of centromeric regions and pangenomic approaches will be needed. It will also be important to study how these genetic effects manifest in different blood cell types and in other tissues, and whether they can influence cancer risk or errors in meiosis.
The results will affect the genetics of aging, oncology and population genomics, as well as the interpretation of clinical findings on mosaic chromosomal abnormalities.
The next step will be functional characterization of the identified variants, especially in kinetochore genes and centromeric repeats, using cell models and genome editing.
The work is directly connected to unresolved questions about why different chromosomes have different propensities for missegregation during cell division and how inherited variants in centromeres and kinetochores affect genome stability. This is a key problem in cell biology and genetics, with implications for understanding age-related diseases and cancer.
One of the structural variants discovered, a hybrid chromosome 13/22, arose from a rare recombination between centromeres and occurs in roughly one in 30 people of southern European ancestry.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.