Human de novo mutation rates from a four-generation pedigree reference.

Area of Science:

• Genomics
• Human Genetics
• Molecular Biology

Background:

• Accurate estimation of human de novo mutation (DNM) rates is crucial for understanding genetic variation and disease.
• Previous studies were limited by incomplete genome sequencing and phasing capabilities.

Purpose of the Study:

• To comprehensively characterize the spectrum and rate of human DNMs using advanced sequencing technologies.
• To investigate the influence of genomic features, such as repeat content, on mutation rates.
• To determine the paternal bias and postzygotic origins of germline DNMs.

Main Methods:

• Utilized five complementary short-read and long-read sequencing technologies to phase and assemble over 95% of diploid human genomes.
• Analyzed a four-generation, twenty-eight-member family (CEPH 1463) for de novo variant detection.
• Integrated high-resolution recombination maps to correlate mutation events with meiotic crossovers.

Main Results:

• Estimated 98-206 DNMs per transmission, with specific rates for single-nucleotide variants, indels, structural variants, and centromeric mutations.
• Identified short tandem repeats and variable-number tandem repeats as highly mutable, with 32 loci showing recurrent mutations.
• Demonstrated that DNM rates vary significantly based on repeat content, length, and sequence identity, with a strong paternal bias (75-81%) for germline mutations.
• Estimated 16% of de novo single-nucleotide variants to be postzygotic in origin, lacking paternal bias.

Conclusions:

• The study provides a high-resolution, near-telomere-to-telomere familial genome dataset as a truth set for understanding human genetic variation.
• Genomic repeat content is a major determinant of mutation rates, and a substantial fraction of mutations arise postzygotically.
• Despite a strong paternal bias in germline mutations, postzygotic events contribute significantly to de novo variation without this bias.