Why Are South Asians Missing From Global Health Databases, Pg II

Key Highlights:

• Over one in ten adults globally live with diabetes, with South Asians facing a higher and earlier risk.
• Less than 1% of participants in global Genome-Wide Association Studies (GWAS) are of South Asian ancestry, despite comprising over 20% of the world's population.
• This lack of diversity in health databases, including U.K. Biobank, leads to less accurate diagnostic tools and treatments for South Asian populations.
• Initiatives like the GenomeIndia Project aim to address this gap by building comprehensive genetic databases for the region.

Detailed Insights:

• The underrepresentation of South Asians in genomic and health datasets hinders the effectiveness of AI and machine learning tools in healthcare.
• Polygenic risk scores, which estimate genetic disease risk, are less accurate for South Asians when developed using predominantly European data.
• Diagnostic thresholds and prediction models developed from European populations often require recalibration for South Asian populations due to genetic variations.
• South Asia exhibits immense genetic diversity, with significant differences even within countries like India, making a "one-size-fits-all" approach ineffective.
• Historical funding disparities have led to inadequate research infrastructure and biobanking facilities in low- and middle-income countries.
• Regional collaborations among existing biobanks and cohorts are proposed to build a robust data infrastructure and ensure equitable scientific benefits.

Scientific/Technical Concepts Involved:

• Biobanks: Repositories storing biological samples and associated health data for research.
• Genomic Data: Information derived from an organism's complete set of DNA, including genetic variants.
• Polygenic Risk Scores: A measure combining the effects of multiple genetic variants to estimate an individual's risk for a disease.
• Genome-Wide Association Studies (GWAS): An observational study of a genome-wide set of genetic variants in different individuals to see if any variant is associated with a trait.