Understanding Polygenic Risk Scores: What They Are, Who They Work Best For, and How They Compare to Single-Gene Testing

Genetics is no longer just about single faulty genes. Scientists can now look across your entire genome and add up thousands of tiny signals to estimate your risk for common diseases. This article covers three things: what a polygenic risk score (PRS) actually is, why these scores work better for some people than others depending on ancestry, and how PRS compares to the single-gene tests you may already have heard of.

What Is a Polygenic Risk Score?

The word “polygenic” simply means “many genes.” Most cases of common conditions, such as heart disease, type 2 diabetes, and certain cancers, are not caused by one broken gene. Instead, hundreds or even thousands of genetic variants each nudge your risk up or down by a tiny amount. A polygenic risk score adds all those nudges together into a single number.

To build a PRS, researchers use results from a genome‑wide association study, or GWAS (pronounced ‘gee‑wass’). In a GWAS, scientists compare the DNA of hundreds of thousands of people who have a disease with the DNA of people who do not. They look for genetic variants that appear more often in the affected group, and they record how strongly each variant is linked to the disease. Those weights are then used to calculate a score for any new individual.

The score shows how your inherited genetic risk compares with that of people in a reference population. It may be reported as a percentile or a z-score. A percentile shows how your score ranks among others, for example, the 90th percentile means your score is higher than 90% of the reference population, not that you have a 90% chance of developing the condition. A z-score shows how far your score is above or below the average: positive values are above average and negative values are below average.

It is not a diagnosis, and it is not a certainty. Think of it like a weather forecast: a 70% chance of rain means rain is more likely, not guaranteed. A high PRS means your genetic background puts you at elevated risk, but lifestyle, environment, and chance all still play a role.

PRS is used in two main settings. In some clinical and research settings, it can help doctors identify patients who would benefit most from earlier screening or preventive treatment. In consumer genomics (direct-to-consumer DNA tests), it gives individuals a snapshot of their inherited risk profile.

The Ancestry Portability Problem

Here is where PRS runs into a significant limitation. Historically, GWAS have included far more data from people of European ancestry than from other ancestry groups. When a risk score is built mostly from one group’s data, it tends to be less accurate when applied to people from different backgrounds.

Imagine learning to predict the weather using only data from London. Your model would get quite good at forecasting British weather, but it would struggle in Mumbai or Lagos, where the climate patterns are different. PRS faces a similar challenge: the genetic variants and their effect sizes identified in European populations do not always translate cleanly to people of African, East Asian, South Asian, or admixed ancestries.

This is not a minor technical footnote. It has real health equity consequences. If PRS tools are less reliable for people of non-European ancestry, those individuals may receive less accurate risk estimates, potentially missing out on early interventions or receiving misleading reassurance. People who already face barriers to healthcare access could be further disadvantaged by tools that were not built with them in mind.

Researchers are actively working to close this gap. Large, diverse GWAS cohorts are collecting genetic data from underrepresented populations. Scientists are also developing multi-ancestry PRS methods that combine data from multiple populations to produce scores that perform more equitably across different groups. Progress is being made, but the field still has significant work ahead.

How PRS Differs from Single-Gene Testing

Single-gene testing looks for a specific, high-impact variant in one gene. Monogenic disorders are conditions where a single faulty gene variant is enough, on its own, to cause or very strongly predispose someone to a disease. Classic examples include variants in the BRCA1 and BRCA2 genes, which substantially raise the risk of breast and ovarian cancer, and variants in the LDLR gene, which cause familial hypercholesterolemia (a condition where cholesterol builds up to dangerous levels from birth).

These single-gene variants are rare in the general population, but their effect is large and relatively predictable. If you carry a pathogenic BRCA1 variant, your lifetime risk of breast cancer is dramatically higher than average, and that finding alone can guide major clinical decisions.

PRS works very differently. Instead of one powerful variant, it aggregates hundreds or thousands of common variants, each with a very small individual effect. No single variant in a PRS is decisive on its own. The risk is probabilistic and sits on a continuous spectrum, and it interacts with factors like diet, exercise, smoking, and body weight in ways that a monogenic result typically does not. Monogenic risks can also be modified by polygenic background, lifestyle, treatment, and other factors.

The two types of testing are not competitors. They answer different questions. Monogenic testing asks: “Do you carry a rare, high-impact variant that strongly predicts disease?” PRS asks: “Where do you sit on the population-wide spectrum of inherited risk for this common condition?” Used together, they give a more complete picture. A person could have a low PRS for heart disease but still carry an LDLR variant that puts them at high risk, or vice versa.

Understanding both tools, and their limitations, helps patients and clinicians make better-informed decisions.

Genetics is a rapidly evolving field, and polygenic risk scores represent one of its most promising frontiers. They are not crystal balls, and they may not be equally reliable for everyone yet. But as datasets grow more diverse and methods improve, PRS may become an increasingly useful part of personalized medicine for people of all backgrounds. Knowing what these scores can and cannot tell you is the first step toward using them wisely.