Why Ethnicity Matters in DNA Health Testing: Ancestry Bias in Genetic Reports
78% of GWAS participants are European, but only 16% of the world is. Polygenic risk scores are less accurate for non-European ancestries. Here’s what that means for your results.
Category: Diagnostics · By GenomeNarrator Team · · 6 min read
One of the most important limitations in modern genomic medicine is a systematic bias in the research base: most large genetic studies have been conducted in people of European ancestry.
The diversity problem by the numbers
A 2019 analysis found that 78% of participants in published GWAS studies were of European ancestry, despite Europeans representing less than 16% of the global population. This has real consequences: polygenic risk scores trained on European-ancestry cohorts perform substantially worse in African, East Asian, and South Asian populations.
Why this matters for your results
A risk score that was validated in a European-ancestry population may overestimate or underestimate your risk if your ancestry differs. Variants that appear "rare" in European populations may be common in yours, and vice versa, which affects both how a variant is classified and what its associated risk actually is.
What GenomeNarrator does
We annotate each risk finding with the population in which its association was established. Where multi-ancestry data exists, we use it. We are transparent about which scores have limited cross-ancestry validation so you can weight those results accordingly.
The research is catching up
Major initiatives like the All of Us Research Program, the Uganda Genome Resource, and the China Kadoorie Biobank are substantially expanding the diversity of genomic databases. The field is improving, but it's important to be honest about where the gaps currently are.
More from GenomeNarrator: all genomics articles, how the analysis works, or GenomeNarrator home.