Ancestry and Population Applicability of Genetic Findings

The core problem

Genetic research has not studied all populations equally. Most findings used in tests today come from large studies that mostly enrolled people of European ancestry, so a result may be more accurate for some groups than for others. This page explains what studies found. It is information only and not medical advice. It does not say that ancestry decides anyone's health. "Ancestry" here describes which reference data a method was built on.

Polygenic scores

A polygenic risk score adds up the effects of many variants to estimate relative risk for a common condition. NHGRI notes that most genomic studies to date examined people of European ancestry, so data for other populations may be inadequate, and that the accuracy of these scores may only be valid and useful for European ancestry populations at this point. [1] A review in Nature Genetics concluded that available scores are several times more accurate in individuals of European ancestry than in other ancestries, and that this follows from the Eurocentric bias of genome-wide association studies. [2]

A separate analysis looked at the first decade of polygenic scoring studies (2008 to 2017). It found that 67% included only European-ancestry participants, and only 3.8% involved cohorts of African, Hispanic or Indigenous peoples. [3] The same analysis found lower predictive performance of European-derived scores in non-European samples. [3] See polygenic risk scores.

Variant classification

The same imbalance affects variant reports. In germline testing, variants from understudied ancestries have been disproportionately classified as being of uncertain significance. [4] That study also found reclassification rates differed by ancestry group. [4] Read more in variant interpretation and uncertain significance.

Ancestry tests and their reference databases

Consumer ancestry estimates compare a person's DNA to reference databases of people with known ancestry. MedlinePlus Genetics notes that different companies use different reference databases and methods, so estimates may not be consistent from one company to another. [5] It also says populations may not be equally represented in a given database, so results for some populations may be less informative than for others. [5] Results may also reflect where a family originated rather than where its members lived. [5]

What this means for wellness products

• A "genetic health" or "gene-based wellness" report built on a score validated mostly in one ancestry group may not mean the same thing for you.
• Claims of "personalized" accuracy are marketing claims unless the company shows performance in people like you. See claim check: personalized means better.
• Even where a score is more accurate, it only shows relative risk and always gives probabilities, not certainties. [1]

What is being done

The authors of the Nature Genetics review said that greater diversity must be prioritized in genetic studies and that summary statistics must be publicly shared, noting that early efforts to diversify show promise. [2] Whether this will close the gap is not yet known.