GWAS

We used regenie for the FinnGen R12 release. Regenie's main advantages are fast leave-one-chromosome-out relatedness calculation which avoids proximal contamination, and use of an approximate Firth test which gives more reliable effect size estimates for rare variants.

Regenie version 2.2.4 was used for the majority of endpoints. Regenie version 3.3 was used for endpoints that did not converge under regenie 2.2.4.

Links:

• regenie preprint

• regenie GitHub repository

• FinnGen regenie GitHub repository

• FinnGen regenie pipeline GitHub repository

We analyzed:

• ​2,502 endpoints

  • 2,499 binary endpoints

  • 3 quantitative endpoints (HEIGHT_IRN, WEIGHT_IRN, BMI_IRN)

• 500,348 samples

  • 282,064 females

  • 218,284 males

• 21,311,644 variants

We included the following covariates in the model: sex, age, 10 PCs, Finngen chip version 1 or 2 , and legacy genotyping batch.