Release of Polygenic Risk Score methodology - What does it mean to our users?

Release of Polygenic Risk Score methodology - What does it mean to our users?

Written by: Edin Hamzic, Chief Science Officer
Grammarly checked by: Jamie Fernandez, Content Specialist

With the latest release, we introduced polygenic risk score (PRS) methodology as the basis of our genetic reports. This update replaces the previous mainly single gene-centric reports with the far more accurate PRS methodology. As our user, you might ask what the difference is. Well, a couple of things will change, and we will explain each of those.

By introducing a new PRS methodology to estimate the genetic predisposition to various genetic traits and disorders, BioCertica is implementing the latest methodology in the fast-paced field of genomics.  

Up to this point, BioCertica used a set of points of genetic variations known as SNPs (pronounced snips) to evaluate the genetic predisposition for 75 traits that are assessed in BioCertica’s  products. More about how we selected SNPs and how we estimate genetic risk up to now can be found here and here. This estimation was based on the limited number of SNPs (~300 SNPs) and assumed that all genetic variants contribute equally to their associated trait.

With PRS methodology, we are moving from 300 SNPs to over 40,000 SNPs to estimate the genetic risk for our 75 traits. To put it in simple terms: the more SNPs, the better resolution, and consequently, the better results. To make it more clear, you can compare this with going from a regular HD TV screen to a 4K TV screen. But, it is not only about the number of SNPs; it is also about making a better assessment of how much each SNP influences a given trait.

One might say not all SNPs are born equal, meaning that some have a stronger effect in causing the onset of a given disorder while others have a weaker effect in causing the onset of the same disorder. This is the foundation of PRS methodology. If you would like to read more about PRS methodology and how it works, please check our article on the topic here.

Why is it important for our users?

So, you might ask yourself why is this important for you as a BioCertica user? Well, by increasing the number of SNPs and improving genetic risk score methodology estimation, your results will change accordingly in one or other direction. That is why we want to inform you about this and explain why this is happening.

How will the results change? The best way to reason and explain why this is happening is by going through an example. We will use an arbitrary example.

Up to now, to evaluate the risk score for one's ability to manage obesity, we used two SNPs that have been retrieved from scientific publications:

SNP ID

Genotypes with risk alleles)

Trait

rs12970134

A (1)

Ability to manage obesity

rs17782313

C(1)

Ability to manage obesity

In this case, each allele (column named Genotypes with risk alleles) has been assigned equal effect either in a positive or negative direction. These are summed up and normalized, and this is how we obtained the relative risk. For example, you can see that each allele was assigned the effect value of 1; therefore, if a person is heterozygous (AG, they would have the effect value of 1; however if a person were homozygous for A (AA), they would have the effect value of 2. In this case, our relative risk for the ability to manage obesity was based on the effect of these two SNPs.

However, now with the PRS methodology, we have many more SNPs. Actually, in the case of the ability to manage obesity, we have 113 SNPs at our disposal to estimate the risk for this trait. Moreover, each of these SNPs has a unique effect weight that is estimated based on peer-reviewed scientific literature and retrieved from PGS Catalog or in some cases, we have to build our effect weight estimates based on the GWAS summary files from GWAS Catalog. These unique effect weights are used to estimate the relative risk score. Moreover, we have information on what populations and ancestries these score weights have been estimated and if they fit better to your genetic makeup.

SNP id

Genotypes with risk alleles)

Trait

rs9564267

A (0.023)

Ability to manage obesity

rs1441264

A (0.027)

Ability to manage obesity

rs2389316

C (-0.032)

…an additional hundred SNPs that can’t be fitted here…

…an additional hundred SNPs that can’t be fitted here…

…an additional hundred SNPs that can’t be fitted here…

rs326217

C (0.021)

Ability to manage obesity

rs4584644

T (0.027)

Ability to manage obesity

So, going from 2 SNPs to 113 SNPs and having a higher resolution of each SNPs effect will consequently change the final results. We see the implementation of PRS as a big methodological improvement that will move us toward a more precise estimation of relative genetic risk. Genomics is a fast-paced discipline, and new genetic variants associated with given traits are discovered daily. Therefore, we strive to keep pace, stay on top of the latest scientific discoveries, and continuously improve our products by utilizing those latest discoveries.

Back to blog

Check out some of our DNA kits below.

1 of 3