Supplementary Figure 2: Illustration of the expression and genotype data sets.
From: Quantification of private information leakage from phenotype-genotype data: linking attacks
Variant genotype dataset contains the genotypes for q eQTL variants for nv individuals. jth entry for kth eQTL is denoted by vk,j . Similarly, the expression dataset contains the expression levels for q genes. The kth expression level for jth individual is denoted by ek,j. The variant genotypes for kth variant are distributed over samples in accordance with the random variable Vk. Likewise, the expression levels for kth gene is distributed per random variable Ek. These random variables are correlated with each other with correlation coefficient, denoted by p(Ek, Vk) (right).
