Twenty genetic loci as discovered in the SHIP-0 data set and their most significant associations to non-targeted metabolic traits.

a alternative candidates for metabomatching exist (S1 Fig)
b additional candidates match to other non-targeted traits that also associate with SNP (S1 Fig)
Chr/Position: Chromosomal location of the lead SNP according to the human reference genome (GRCh37). EA/EAF: Effect allele and frequency. Trait or pairwise ratio: Tested metabolic trait (chemical shift). In case of ratios, the trait that drives the association (i.e., shows the stronger association signal) is named first. N: Number of samples where both genotype and phenotype data were available for the tested SNP/metabolic trait pair. beta’: beta’ is defined as 10beta-1 where beta depicts the relative effect size representing the slope of the regression line in the linear model when using log10-scaled metabolic traits and the occurrence of the SNP’s minor allele (coded as 0,1, and 2). Thus, beta’ describes the relative difference per minor allele copy for non-scaled metabolic traits in comparison to the estimated mean of the metabolic trait in the major homozygote test subjects. P-gain: Defined as min(P(M1)/P(M1/M2), P(M2)/P(M1/M2)), where M1 and M2 represent the two traits of which the ratio M1/M2 is built. Metabomatching: Annotation of the non-targeted metabolic trait at the given chemical shift as suggested by metabomatching (S1 Fig). Replicated: SNP/metabolic trait pair was replicated in KORA F4 (P < 1.32×10−3) (S2 Table). Targeted: SNP or proxy in LD is also associated with a targeted metabolic trait (P < 3.25×10−12) (Table 1).
Twenty genetic loci as discovered in the SHIP-0 data set and their most significant associations to non-targeted metabolic traits.