Inequality in genetics and genomics is an important global issue. For example we need global genomic representation to accurately predict who will develop diseases. At the current moment, the majority of association studies have used European ancestry population datasets. This issue is evident in the results below, when we compared polygenic risk scores for type 2 diabetes of the same individual but testing against two different ancestries. 

By using the data from which ancestry would have communicated the lower risk of developing type 2 diabetes?