These files contain the summary statistics of the meta analyses performed in Draisma, H., Pool, R., Kobl, M. et al. Genome-wide association study identifies novel genetic variants contributing to variation in blood metabolite levels. Nat Commun 6, 7208 (2015) (https://doi.org/10.1038/ncomms8208).
In this directory you can find the meta-analyses outputs per metabolic trait, generated by METAL. Format: tab-delimited.
Legend to column names:
- MarkerName: Name (rsid) of marker;
- Allele1: Allele 1;
- Allele2: Allele 2;
- Freq1: Frequency of allele 1;
- FreqSE: Standard error of the above frequency;
- MinFreq: Frequency minumum;
- MaxFreq: Frequency maximum;
- Effect: Effect size (beta);
- StdErr: Standard error of effect size;
- P-value: P value based on (Effect/StdErr)^2 statistic;
- Direction: Direction of effect (+,- or ? denoting positive, negative or missing effect, respectively) [for cohort input order, c.f. file "MACohortInputOrder.csv"];
- NTot: Total sample size for this marker;
- HetISq: Heterogeneity I^2 parameter;
- HetChiSq: Heterogeneity Chi^2 statistic;
- HetDf: Heterogeneity degrees of freedom;
- HetPVal: Heterogeneity P value, based on above Chi^2 statistic.
Note(1):
Meta-analyses were performed after pre-filtering on HWE P value > 1.0e-6, excluding results that strongly deviate from HWE.
Note(2):
These files have been post meta-analysis filtered on the following criteria:
- HetDF > 0: HetDf==0 means that only for only one cohort the meta-analysis has been performed, hence this is by definition not a meta-analysis result;
- HetISq < 75: This filters out markers of high meta-analysis heterogeneity.
Note(3):
File "MACohortInputOrder.csv" contains the meta-analysis cohort input order for each metabolite. Format: comma-separated.
