Why is age and sex added as covariates in GWAS regressions?

Question

[FeedMeDumplings]

As I understand you add covariates in your model to control for potential confounding effects of the covariates on your exposure (X) and outcome (Y).

So does this mean that age and sex can affect allele frequencies?

Thanks for the wisdom!

Answer 1

[Kevin Blighe]

No, it's that they may affect the outcome, y.

cancer ~ SNP1

SNP1 statistically significantly increases one's risk of cancer

---

---

cancer ~ SNP1 + age

Independent of age, SNP1 statistically significantly increases one's risk of cancer

---

---

cancer ~ SNP1:age

This will reveal in which age groups there is a statistically significant association, assuming that age is now encoded categorically

Kevin