Before doing any GWAS (genome-wide association study) it is necessary to check for the normality of the phenotypic distribution. If the phenotype is normally distributed only once it is log-transformed, what phenotypic data do I have to use while doing the GWAS? The non-transformed one or the log-transformed?

  • $\begingroup$ How is the significance computed? If the computation assumes that the response variable is normally distributed (which is likely the case), then you need it to be normally distributed (hence log-transformed in your case). $\endgroup$
    – Remi.b
    May 2 '18 at 15:45
  • $\begingroup$ Related ncbi.nlm.nih.gov/pmc/articles/PMC3869493 $\endgroup$
    – CKM
    May 2 '18 at 18:42

Yes, use the log-transformed phenotype. if you want to use a normally distributed error term, and this will only occur if the phenotype is log-transformed, then you must log-transform the phenotype.

The phenotype should be distributed in the same manner as the error term in your regression model of the GWAS.

$pheno = \beta_1 X_D + \beta_2 X_A +\epsilon$

$\epsilon = \mathcal{N}(x | \mu,\sigma)$

Following the central limit theorem, most of phenotypes are normally distributed. However, there are cases where a Bernoulli distribution is correct, and a logistic style regression should be used.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.