Calibrated prediction intervals for polygenic scores across diverse contexts

Abstract

Polygenic scores (PGS) have emerged as the tool of choice for genomic prediction in a wide range of fields from agriculture to personalized medicine. We analyze data from two large biobanks in the US (All of Us) and the UK (UK Biobank) to find widespread variability in PGS performance across contexts. Many contexts, including age, sex, and income, impact PGS accuracies with similar magnitudes as genetic ancestry. PGSs trained in single versus multi-ancestry cohorts show similar context-specificity in their accuracies. We introduce trait prediction intervals that are allowed to vary across contexts as a principled approach to account for context-specific PGS accuracy in genomic prediction. We model the impact of all contexts in a joint framework to enable PGS-based trait predictions that are well-calibrated (contain the trait value with 90% probability in all contexts), whereas methods that ignore context are mis-calibrated. We show that prediction intervals need to be adjusted for all considered traits ranging from 10% for diastolic blood pressure to 80% for waist circumference. Adjustment of prediction intervals depends on the dataset; for example, prediction intervals for education years need to be adjusted by 90% in All of Us versus 8% in UK Biobank. Our results provide a path forward towards utilization of PGS as a prediction tool across all individuals regardless of their contexts while highlighting the importance of comprehensive profile of context information in study design and data collection.

Estimating Disorder Probability Based on Polygenic Prediction Using the BPC Approach

“…It may be estimated using external data to obtain a data-informed prior, such as context-specific lifetime prevalence estimates of individuals seeking health care for a specific disorder in a given hospital. The context may refer to any variable that modifies a disorder’s lifetime prevalence, such as age, sex, or income 37 , meaning any covariate can be incorporated into the prior disorder probability. Alternatively, if no data is available, prior elicitation 38 may be used, where a clinician (or a panel of clinicians) provides a subjective estimate of the prior.…”

Integrative polygenic risk score improves the prediction accuracy of complex traits and diseases

“…There might be more contributing factors that influence the prediction accuracy, such as sample sizes of the PGS panel and polygenicity of the traits, 6 , 32 ancestral consistency between discovery GWAS and linkage disequilibrium (LD) reference panels in PRS methods, 33 ancestry proportions in the discovery GWAS, 33 and cohort-specific contexts. 34 …”

scAI-SNP: a method for inferring ancestry from single-cell data

“…For example, consider polygenic scores that aggregate the contributions of many genetic variants to a trait. The accuracy of polygenic risk scores varies significantly with context, such as age and income, in addition to ancestry 45 . In the cases where context is not recorded, ancestry can act as a proxy for environmental and socioeconomic factors.…”