Study defines average human genome and normal genetic disease risk (opens in a new tab)

In plain English

Using WGS from two large population cohorts (UK Biobank: 490,541; All of Us: 414,830) after variant/genotype QC, the authors report per-individual structural-variant/copy-number burden summaries. They state an average genome contains ~67 autosomal deletions/duplications >10,000 basepairs and that rare autosomal CNVs >1 Mb occur in ~1 in 41 individuals (~2.5%).

Key findings

• Average autosomal deletions/duplications >10,000 basepairs per genome ~67.~67 per genome
• Rare autosomal copy number variants >1 Mb present in ~1 in 41 individuals (~2.5%).~2.5% prevalence (1 in 41)

In plain English

Cross-cohort WGS analysis reports per-individual homozygosity and runs-of-homozygosity (ROH) burden, showing that an average genome has ~32–42% of variants homozygous and approximately 4 unique ROH >1.5 Mb, with marked differences across genetically inferred ancestry groups.

Key findings

• A substantial fraction of variants per person are homozygous.~32–42% homozygous
• Individuals carry multiple long runs of homozygosity.≈4 unique ROH >1.5 Mb per person

In plain English

Cross-cohort analysis of coding variation restricted to MANE Select transcripts to quantify per-individual counts and predicted consequence distributions relevant to monogenic disease assessment. Using WGS from two large cohorts, the authors report per-person counts of coding SNVs/indels (21–26k), the fraction predicted protein-altering (~48%), proportions classified as high-confidence loss-of-function (~0.8%) and those with REVEL > 0.7 (~0.1%). They also report per-genome burdens restricted to GenCC-curated monogenic disease genes: ~1–3 ClinVar pathogenic/likely pathogenic variants, ~2–5 high-confidence LoF variants, and ~50 ClinVar variants of uncertain significance, and note that most ClinVar pathogenic variants are heterozygous in recessive genes consistent with carrier status.

Key findings

• Number of coding exon SNVs/indels per person (MANE Select transcripts).21–26,000 variants per person
• Proportion of coding variants predicted to alter the protein.≈48%

In plain English

Across nearly 1 million WGS (UK Biobank + All of Us) and restricting to rare variants in GenCC-curated monogenic disease genes, the average individual carries a small number of ClinVar-designated pathogenic/likely pathogenic variants and a larger number of variants of uncertain significance, with most ClinVar pathogenic variants observed as heterozygotes in recessive genes (consistent with carrier status).

Key findings

• Average genome contains ~1–3 ClinVar pathogenic or likely pathogenic rare variants within GenCC-curated monogenic disease genes.~1–3 per individual
• Average genome contains ~2–5 high-confidence loss-of-function rare variants within GenCC genes.~2–5 per individual

In plain English

Abstract reports that, depending on parental age, the average genome contains ~63 de novo variants, with ~75–80% of those being paternal in origin.

Key findings

• Depending upon parental age, the average genome contains ~63 de novo variants, of which ~75–80% are paternal in origin.~63 de novo variants; 75–80% paternal

In plain English

Within UK Biobank participants, the authors applied polygenic scores for 28 common diseases and report that the "average person" is in the top 10% of PGS for approximately 3 conditions, and that about 1 in 5 people are in the top 1% for at least one PGS.

Key findings

• "Finally, applying polygenic scores (PGS) for 28 common diseases within UKB, we estimate that the average person is in the top 10% for ~3 conditions, and around 1 in 5 people are in the top 1% for at least one PGS."~3 conditions in top 10% per person; ~20% of individuals in top 1% for at least one PGS

In plain English

The authors interpret their empirical WGS findings to argue that pervasive individual genetic variation has practical implications for policy-makers considering genomic screening programs and for clinicians who counsel families. They also state that the analysed cohorts do not represent many global populations and argue for more inclusive sequencing datasets to support equitable precision medicine and generalizable policy decisions.

Key findings

• The ubiquity of genetic variation in individual genomes implies important considerations for policy-makers evaluating genomic screening and for clinicians providing genetic counselling.
• The continued underrepresentation of many global populations in the analysed cohorts is a limitation and supports a recommendation for more inclusive global datasets to enable equitable precision medicine.