Mapping and characterization of structural variation in 17,795 human genomes

Area of Science:

• Genomics
• Human Genetics
• Bioinformatics

Background:

• Whole-genome sequencing aims to identify all genetic variations, but tools for structural variants (SVs) lag behind those for smaller variants.
• Structural variants, unlike single-nucleotide variants and indels, are challenging to detect and characterize comprehensively.
• Understanding SVs is crucial for a complete picture of human genetic diversity and disease association.

Purpose of the Study:

• To map and characterize structural variants across a large cohort of deeply sequenced human genomes.
• To create the largest whole-genome sequencing-based structural variant resource to date.
• To assess the frequency, impact, and implications of structural variants in the human genome.

Main Methods:

• Utilized a scalable computational pipeline to analyze 17,795 deeply sequenced human genomes.
• Mapped and characterized various types of structural variants, including deletions, duplications, inversions, and translocations.
• Publicly released site-frequency data for the identified structural variants.

Main Results:

• Established the largest whole-genome sequencing-based structural variant resource, including 158,991 ultra-rare SVs.
• Found that individuals carry an average of 2.9 rare coding-altering SVs, impacting 4.2 genes.
• Estimated that SVs account for 17.2% of rare alleles genome-wide, with significant predicted deleterious effects, predominantly from noncoding deletions.

Conclusions:

• Structural variants are a substantial source of rare genetic variation with significant functional impact, comparable to loss-of-function alleles.
• The developed resource and findings will aid in the analysis and interpretation of structural variants in human genetics studies.
• Further research into SVs is essential for understanding genetic disease and human evolution in the era of whole-genome sequencing.