Analysis covariates

This page has been last updated for R13.

Sandbox directory

Analysis covariates are available in the following Sandbox directory:

/finngen/library-red/finngen_R[RELEASE]/analysis_covariates

Data files

The analysis covariate file is a tab-separated, gzip-compressed text file that contains covariate and endpoint data for each sample. The file contains three sets of columns:

• column 1: Sample ID

• columns 2 to N: covariates including principal components, ~200 columns for R13

• columns N+1 to N+1+number of endpoints: individual's phenotype status for each FinnGen endpoint

The covariate file does not contain FinnGen genotypes for individuals with non-Finnish ancestry. For more complete phenotype data see the phenotype files.

Often users will subset this file in R to run their own analyses and/or add additional analysis columns.

Some column descriptions:

Column name	Description
FINNGENID	Sample ID
AGE_AT_DEATH_OR_END_OF_FOLLOWUP	Age of sample at death or end of followup
batch	batch
n_var	Number of genotyped variants
chip	Chip used for genotyping
IS_AFFY	Whether the sample was genotyped using Affymetrix chip
IS_FINNGEN1_CHIP	Whether the sample was genotyped using Finngen v1 chip
IS_FINNGEN2_CHIP	Whether the sample was genotyped using Finngen v2 chip
IS_AFFY_*	Whether the chip genotypes were called using the specified version of the calling algorithm
AGE_AT_DEATH_OR_END_OF_FOLLOWUP2	AGE_AT_DEATH_OR_FOLLOWUP*AGE_AT_DEATH_OR_FOLLOWUP
BATCH*	Whether the sample was part of that genotyping batch. Can be used to control for batch-specific effects in analysis.
PC*	Individual's PCA value for that component
*_IRN	Inverse rank-normalized quantitative endpoints

For other columns, refer to the minimum extended phenotype and the endpoint data pages.

Further information

The covariate file is used for GWAS and other analyses. The following covariates are used in FinnGen's core GWAS analyses:

• Age

• Sex

• First 10 principal components

• Genotyping batch (Finngen 1 or 2 chip and legacy genotyping batch)

Note: This file is usually released a little later than the phenotype files as it needs the PCA results to be created.