Phenome-wide association studies across large population cohorts support drug target validation

Summary-level data generated by Genomics plc as presented in:
Diogo, D. et al. Phenome-wide association studies across large population cohorts support drug target validation. Nat. Commun. 9, 4285 (2018). https://doi.org/10.1038/s41467-018-06540-3

If you have any questions or comments regarding these files, please contact Genomics plc at research@genomicsplc.com

NOTES
-----------------------------
These analyses were carried out using the interim UK Biobank imputation data release. Analyses were restricted to a subset of "white-British" unrelated samples with a maximum sample size of 112,337 individuals. 

Case control phenotypes were defined based on categorical datafields as listed in the accompanying file. 
Quantitative phenotypes were either rank-normalised before analysis, or beta/se values were standardised after analysis using the variance of the phenotype. The normalisation value is indicated in the accompanying file.
 
All analyses included Age at assessment, sex, genotyping chip, and 10 principal components as covariates. 

We used plink1.9 linear/logistic regression as appropriate. For chromosome X variants males were treated as having 0 or 2 alternative alleles. 

The results are not adjusted for genomic control.

DATA FILE CONTENT DESCRIPTION
-----------------------------
CHR - Chromosome
SNP - Variant rsID
ALT - Alternative allele (effect allele)
REF - Reference Allele (non-effect allele)
BP - Position in base pairs (b37, 1-based)
NMISS - Number of samples with non-missing genotypes
BETA - Effect size (log odds ratio or standardised effect size)
SE - Standard error
P - P-value
F_MISS - genotype missing rate
P_hwe - Hardy-weinberg p-value
MAF - ALT allele frequency