Rethinking race, ethnicity and ancestry in genetics and genomics research

[CharonY]

Recently we had a discussion on nature vs nurture in this forum and I have mentioned the difficulties of looking at complex traits as directly and firmly genetically linked. Specifically I mentioned the misuse of race and ethnic groups in this context. Now I cam a cross a publication of the National Academies which elaborates on this issue:

https://nap.nationalacademies.org/26902

A summary can be found here: https://www.nationalacademies.org/news/2023/03/researchers-need-to-rethink-and-justify-how-and-why-race-ethnicity-and-ancestry-labels-are-used-in-genetics-and-genomics-research-says-new-report

 

Quote

Almost all human traits are a result of the interplay between genetic and environmental factors. Instead of relying on population descriptors as proxies for describing the effects of environmental factors, researchers should incorporate environmental factors in their work, and use variables that capture more precise information, the report says. Genetics and genomics researchers should collaborate with experts in the social sciences, epidemiology, and other disciplines — as well as work in partnership with communities — to aid in these studies. 

Quote

hanging Research Practices

The report says researchers should:

• Not use race as a proxy for human genetic variation. In particular, they should not assign genetic ancestry labels to individuals based on their race, regardless of whether the label was self-identified.  
• Apply labels consistently to all participants. For example, if ethnicity is the most appropriate descriptor, all participants should be assigned an ethnicity label, rather than labeling some by race and others by ethnicity.  
• Be attentive to the connotations and impacts of terminology they use to label groups. The report points to the term “Caucasian” as an example, explaining it should not be used under any circumstance because it was originally coined to convey the notion of white supremacy.  
• Disclose the process by which they select and assign group labels. If researchers develop new labels for existing samples, researchers should provide a description of the differences between the new and old labels.  

Much of it is a call for more precision, but also highlight the harm of misconceptions which can arise from sloppy work.

[Trurl]

Sounds dangerous to use race and ethnic labels. But what if in studying the genome there are more “classifications?”

These classifications may be rooted in science but would also be misused.

[CharonY]

5 minutes ago, Trurl said:

Sounds dangerous to use race and ethnic labels. But what if in studying the genome there are more “classifications?”

These classifications may be rooted in science but would also be misused.

Can you provide an example? The report focuses mostly on race and adjacent factors as there is a well-documented history of issues. The general thrust of the argument is to incorporate more detailed measurable parameters rather than pre-classifications.

So a study would try to factor in things like income or perhaps other well-studied wealth indicators, rather using a more vague classifier such as "working class". Unless, of course it is well defined via other parameters. There may also be good ways to incorporate ethnicity, it is just that the practice has been highly problematic resulting in non-reproducible outcomes (in good scenarios) to devastating misunderstandings and resultant harm in medical practice.