What Are Our AIMs? Interdisciplinary Perspectives on the Use of Ancestry Estimation in Disease Research

Abstract

Background: Ancestry estimation serves as a tool to identify genetic contributions to disease but may contribute to racial discrimination and stigmatization. We sought to understand user perspectives on the benefits and harms of ancestry estimation to inform research practice and contribute to debates about the use of race and ancestry in genetics. Methods: Key informant interviews with 22 scientists were conducted to examine scientists’ understandings of the benefits and harms of ancestry estimation. Results: Three main perspectives were observed among key informant scientists who use ancestry estimation in genetic epidemiology research. Population geneticists self-identified as educators who controlled the meaning and application of ancestry estimation in research. Clinician-researchers were optimistic about the application of ancestry estimation to individualized risk assessment and personalized medicine. Epidemiologists remained ambivalent toward ancestry estimation and suggested a continued role for race in their research. Conclusions: We observed an imbalance of control over the meaning and application of ancestry estimation among disciplines that may result in unwarranted or premature translation of ancestry estimation into medicine and public health. Differences in disciplinary perspectives need to be addressed if translational benefits of genetic ancestry estimation are to be realized.

Keywords:

• bioethics
• continental population groups
• genetics
• interdisciplinary communication
• translational research

Acknowledgments

This work was supported by the National Science Foundation (Doctoral Dissertation Research Improvement Grant SES0822410 to Dr. Joon-Ho Yu) and the National Institutes of Health (P50 HG003374 to Dr. Wylie Burke). The authors thank Kelly Edwards, PhD, and the anonymous reviewers of AJOB Primary Research for their critical review of this article.

Notes

A marker may have one or more states (e.g., Marker 1 could be either A or B or C). To infer ancestry of an individual, all markers are tested and then statistical algorithms are used to analyze the results to determine the probable ancestral origins of the genome. This is done by comparison to other datasets of marker data from “ancestral” reference populations.