SPLENDID incorporates continuous genetic ancestry in biobank-scale data to improve polygenic risk prediction across diverse populations.

SPLENDID improves polygenic risk prediction accuracy for non-European and admixed populations without requiring ancestry labels in large-scale biobank data.

• Why it matters: Accurate disease risk prediction across diverse ancestries is limited by current methods that depend on ancestry labels, which are impractical in clinical settings and may contribute to health disparities.
• What they did: The authors developed SPLENDID, a penalized regression framework that models genetic ancestry as a continuum, applied to individual-level data from over 564,000 participants in the All of Us Research Program and UK Biobank.
• The result: SPLENDID significantly outperformed existing methods in prediction accuracy, especially for non-European and admixed groups, enabling more equitable and robust genetic risk assessments.