Skip to content
PaperFren

Do polygenic scores work equally within one ancestry?

Open paper intelligence

Even within a relatively homogeneous ancestry group, PGS prediction accuracy differs substantially across socio-economic and related strata.

Source

Variable prediction accuracy of polygenic scores within an ancestry group

Mostafavi H, Harpak A, Agarwal I, et al. · eLife · 2020

doi.org/10.7554/elife.48376Read the full paper ↗400 citationscc by

Study at a glance

Design
Computational / modelling — Within-ancestry stratification of PGS accuracy in UK Biobank (education, height, BMI, etc.)
N
N=408434 · 408,434 UK Biobank participants passing QC; analyses often within White British strata
Population
UK Biobank participants of broadly similar genetic ancestry
Outcome
Within-ancestry variation in polygenic score prediction accuracy

Structured fields used in claim comparison tables when every cited study has a complete layer.

What they did

Evaluated PGS prediction accuracy across strata in UK Biobank individuals of similar ancestry, focusing on traits like education, height, and BMI.

What they found

Major within-ancestry differences in PGS accuracy among groups with similar genetic ancestry; accuracy depends on ascertainment/SES-related structure, not ancestry labels alone.

The limits

What it doesn't show

Does not claim PGS are clinically actionable for all traits.

Key terms

Polygenic score (PGS)
Sum of GWAS allele effects used for phenotypic prediction.
Portability
How well a score predicts in a new sample/group.
SES
Socio-economic status.
UK Biobank
Large human cohort used for genetic analyses.

Flashcards

1 / 11

Research intelligence for this paper

See its role on concept claims, tensions it is part of, placement history, and related discoveries.

Open paper intelligence

Quiz yourself

1 / 6

Main claim

Common questions

Only a cross-ancestry problem?

No—accuracy also varies within ancestry.

Cohort?

UK Biobank homogeneous-ancestry subset.

Example traits?

Education, height, BMI.

Why?

Ascertainment/SES-related structure and related factors.

More on GWAS