How we estimate genetic risk
Our calculations of your estimated lifetime risk for each condition are based on:
The average lifetime risk that a person of your gender will have that condition.
This is compiled from authoritative epidemiological reports in the medical literature. The figures are based on the total lifetime risk for that condition, typically for a U.S. population and for your gender.
The risk for some rare conditions is difficult to know with certainty and is estimated from the best available published evidence, using information about the incidence of the condition and the population age distribution in the United States.
For some conditions, such as hemochromatosis and lactose intolerance, lifetime risk estimates are not available.
Which genetic risk markers we found when we analyzed your genetic makeup.
These genetic risk markers meet our criteria for reliability and significance. The tested risk markers are listed within your genetic test results for each condition. For each health condition, we test your DNA sample for one or more SNPs (single nucleotide polymorphisms) that have been shown to be associated with the health condition. We select these SNPs based on the strength and reproducibility of the evidence reported in the scientific literature. (If you want to read this literature, we provide the citation for each marker we use.)
How much your genetic risk markers increase your odds of having the health condition, compared to someone with no markers.
These "odds ratios" are based on the original studies that identified these genetic risk markers as being associated with disease risk. From the scientific literature, we have obtained the odds ratio showing the increase in the risk of a condition among subjects who had one or two copies of the genetic risk marker at each SNP location. The odds ratio depends on the number of copies of the risk marker. Some studies only report the odds ratio for a single marker, and in those cases we make some assumptions to calculate the odds ratio for two genetic risk markers.
For some conditions, such as hemochromatosis and lactose intolerance, odds ratios are not available for individual SNP markers. For hemochromatosis, this is because the three SNPs do not contribute to risk independently, but rather act in combination to affect risk. For the SNP associated with lactose intolerance, the association is so strong that the odds ratio is infinite. In other words, in the study population, no cases were seen with the non-risk genotype, nor were any controls seen with the risk genotype. For the medications outcomes, odds ratios are not presented. Instead, we indicate whether you have increased risk compared to the rest of the population.
A model for visualizing, ranking and estimating lifetime risk by combining genetic risks when we test for more than one marker for a condition.
- We use a "multiplicative" model for combining the odds ratios from each genetic risk marker in our scan. To estimate the combined effect on a person's risk, we have applied a formula that takes into account the lifetime risk of that condition and the prevalence of each marker within a reference population; converts the odds ratio to a relative risk; and finally multiplies the risks together.
- We also compare you to one of the reference populations in the International HapMap Project, a large publicly available data set. As additional reference population databases become available we will update this information, and you may find the lifetime risk numbers become more precise based on different ethnicities.
- To determine your estimated lifetime risk, we take into account the lifetime risk of the health condition and the average score in the reference population. This formula assumes that marker frequencies in the reference population are representative of your own ancestry, that each marker has an independent effect on your overall risk and that the distributions of the marker versions in the general population are independent.
- For conditions where lifetime risks and odds ratios are not available, such as hemochromatosis and lactose intolerance, we do not use this model, but rather report whether you have increased risk, compared to the rest of the population.
The result is an estimate of your lifetime risk for that health condition, based on the markers we found in your genome, which you can then compare against the average risk for that condition.
It is an estimate because the above assumptions, while reasonable, may not always be true. We also tell you where your genetic risk factors rank, compared to the rest of the population.