Statement on Ancestry

Genetic analyses have predominantly been conducted on samples of European ancestry (as defined by genetics). This emphasis on a specific population group has arisen for several reasons, including data availability, the desire to study homogenous groups in genome wide association studies (to limit bias due to population stratification), research resources (e.g. the availability of funding, expertise, technology, and infrastructure needed to collect, process, and analyse genetic data) and the underrepresentation of some population groups in genetic research studies.

It is important to acknowledge the limitations of this. For example, results solely examining European ancestry groups may not generalise to other ancestral groups and therefore the superpopulation of interest. This includes analyses which are increasingly common in the research community such as Genome-wide Association Studies and the use of polygenic scores, which may be less informative and/or more biased when used in non-European ancestry groups.

This is particularly important when analysing ethnically diverse populations. If researchers limit to European ancestry in CLS studies we recommend that they acknowledge this potential limitation in their DAC applications and research papers; and provide brief justification. For example:

“Our study was limited to those of European ancestry (delete as appropriate: in order to study a homogeneous group and limit bias / to aid comparability with previous studies / given less diverse cohorts / since comparable polygenic scores were not available in all ancestral groups). This may limit the generalisability of our findings to the total UK population. As genetic evidence becomes increasingly available in broader ancestral groups, future research should extend these findings to other groups.”

CLS is making efforts to address this issue. The MCS oversampled underrepresented groups and these study members can be included in ongoing multi-ancestry initiatives to increase diversity and participation in genetic research. We are exploring the feasibility of creating polygenic scores for all cohort participants regardless of ancestry.