From: Implementation of individualised polygenic risk score analysis: a test case of a family of four
2a. PRS design selection criteria | 2b. Bioinformatic selection criteria |
---|---|
i. Traceable to a discovery GWAS from a recognised consortium | i. At least 95% SNPs present in 1000G Phase III distribution |
ii. Trained in a second study using previously published PRS creation methods | ii. Pass matching allele filter |
iii. Validation in a large independent cohort; preferably UKB | iii. Have effect weights |
iv. AUC or similar performance metric above 0.55 | iv. Coordinates in hg19 |
v. If several PRS of same phenotype: judgement of study as a whole | Â |
vi. Have risk metrics if possible: odds ratios, hazard ratios, fold increase | Â |