Skip to main content

An introduction to genetic association analysis

Published on August 31, 2026   23 min

Other Talks in the Series: Periodic Reports: Advances in Clinical Interventions and Research Platforms

Please wait while the transcript is being prepared...
0:00
Hi. I'm Mark Iles, I'm an associate professor in statistical genetics at the University of Leeds. I'm going to present you with an introduction to genetic association analysis.
0:12
First, what do we mean by genetic association? Genetic association is the statistical association between a genetic variant and a trait. A trait could be categorical, for instance, whether or not someone has a disease; or it could be continuous, such as height, blood pressure, hormone levels. For a disease trait, if the frequency of the disease varies by genotype or equivalently, genotype frequencies are different in cases and controls, then this is genetic association.
0:40
Let's take an example of genetic association with disease risk. Here we have a single nucleotide polymorphism or SNP, for short. The alleles are A and G. So three genotypes are possible: AA, AG and GG. The prevalence of asthma varies according to genotype. Amongst those with the AA genotype, 325 out of every 100000 have asthma, 450 out of every 100000 for those with the AG genotype and 510 in every 100000 for the GG genotype. Thus, the SNP is associated with a risk of asthma.
1:16
Here we have a similar example but for a continuous trait, in this case, height. We can see the distribution of height varies by genotype. The mean height is lowest in those with an AA genotype and highest in those with a GG genotype.
1:30
Testing for association between a trait and a genetic variant is simple. Again, we have genotypes AA, AG and GG and a binary trait, cases and controls for disease here. Tabulating the numbers of each genotype for cases and controls, we can see that the AA genotype is less frequent in cases than controls, while the AG genotype is slightly more frequent in cases than controls, and the GG genotype is even more frequent in cases than controls. This suggests that the G allele is associated with increased risk of disease. We can test this formally with a chi-squared test. We can also estimate the effect size by calculating the odds ratio for the AG genotype compared to the AA genotype's baseline and the GG genotype with again the AA genotype's baseline. Remember that if the disease is rare, the odds ratio approximates to the relative risk. This confirms that someone with the AG genotype is at increased risk of disease, about 30% higher than someone with the AA genotype, while someone with the GG genotype is at about twice the risk. This test makes no assumptions about the underlying risk. As is usual with statistical tests, making no assumptions is fine. But if you can make assumptions in your modelling that are correct or close to correct, then you can increase your power. If we know or suspect that

Quiz available with full talk access. Request Free Trial or Login.

Show
Hide

An introduction to genetic association analysis

Embed in course/own notes