
South Asians account for less than 1% of participants in global genome-wide association studies between 2005 and 2025, despite representing over 20% of the world's population.
The reliance on European-heavy datasets makes AI-driven health tools and polygenic risk scores less accurate for South Asians, who face higher rates of type 2 diabetes and cardiovascular disease.
India alone is projected to have 125 million people with diabetes by 2045, highlighting the urgent need for more diverse genomic data to tailor medical treatments to the population.