How does deCODEme model the genetic risk?
For diseases with single confirmed SNP, the risk from the published reference is used. It is however important to note that a user might not find the exact relative risk number displayed in deCODEme in the corresponding reference. This is because, in most scientific publications, the relative risks, or odds-ratios, estimated are usually between two alleles or two genotypes. Since the average population risk can be expressed as a weighted average of the risks of different genotypes, using results from the reference, deCODEme performs the appropriate calculations to calculate the risk of a genotype group relative to the population, making it easier for the user to interpret the results. In our risk calculations for diseases with multiple SNPs, we assume that the risk associated with the different loci behaves independently, i.e. the risk is multiplicative in nature. This is a parsimonious model that usually fits the data of complex traits reasonably well. When judged to be well supported
