Skip to content

PRS evaluation

PGS performance was evaluated using two main metrics:

  1. PGS Correlation
    Pearson’s correlation between PGS derived from imputed SNP array data and PGS from whole-genome sequencing (WGS).

  2. ADPR (Absolute Difference in Percentile Ranking)
    The absolute difference in percentile ranking between PGS from array-imputed data and the WGS-derived gold standard.

These evaluations were conducted across multiple p-value thresholds to ensure unbiased comparison, based on the method from Nguyen et al., 20221.

PRS formula

For an individual \(i\), the polygenic score at p-value threshold \(P_{T}\) is calculated as:

\[ PGS_i(P_T) = \sum_{j=1}^{M} {1}_{\{P_j < P_T\}} \, x_{ij} \, \hat{\beta}_j \]
  • \(P_T\) : p-value threshold
  • \(M\) : number of SNPs after clumping
  • \(x_{ij}\) : allele count for SNP \(j\) in individual \(i\)
  • \(\hat{\beta}_j\) : marginal effect size from GWAS for SNP \(j\)
  • \(1_{\{P_j < P_T\}}\) : indicator function for p-value filtering

PGS correlation

  • It is the Pearson correlation coefficient between imputed and true sets of raw PGS values computed for the same individuals.
  • Interpretation: Measures how similar the PGS values are in scale and ranking across two methods.

Absolute Difference in Percentile Ranking (ADPR)

  • It is the average absolute difference in percentile rank of each individual between imputed and true sets of PGS.

  • Formula:

    \[ \text{ADPR} = \frac{1}{N} \sum_{i=1}^{N} \left| \text{percentile}_i^{(A)} - \text{percentile}_i^{(B)} \right| \]
    • N is the number of individuals
    • percentile_i_A and percentile_i_B are the percentile ranks of individual i in each PGS imputed and true distribution

  1. Dat Thanh Nguyen, Trang TH Tran, Mai Hoang Tran, Khai Tran, Duy Pham, Nguyen Thuy Duong, Quan Nguyen, and Nam S Vo. A comprehensive evaluation of polygenic score and genotype imputation performances of human snp arrays in diverse populations. Scientific Reports, 12(1):17556, 2022. ↩