FairPRS: a fairness framework for Polygenic Risk Scores
Machado-Reyes, D.; Bose, A.; Karavani, E.; PARIDA, L.
Show abstract
Polygenic risk scores (PRS) are increasingly used to estimate the personal risk of a trait based on genetics. However, most genomic cohorts are of European populations, with strong under-representative of multi-ethnic minority groups. Given that PRS poorly transport across racial groups, this has the potential exacerbate health disparities if used in clinical care. Hence there is a need to generate PRS that perform comparably across ethnic groups. Borrowing recent advancements in the domain adaption field of machine learning, we propose FairPRS - an Invariant Risk Minimization (IRM) approach for estimating fair PRS or debiasing pre-computed ones. We test our method on both a diverse set of synthetic data and real data form the UK Biobank. We show our method can create ancestry-invariant PRS distributions that are both racially unbiased and largely improve phenotype prediction. We hope that fair PRS will contribute to fairer characterization of patients by genetics rather than by race.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Incorporating family disease history and controlling case-control imbalance for population based genetic association studies 96%
- An exact, unifying framework for region-based association testing in family-based designs, including higher criticism approaches, SKATs, multivariate and burden tests 94%
- Summary statistics from large-scale gene-environment interaction studies for re-analysis and meta-analysis 94%
Similar papers in this journal
- Prioritization of disease genes from GWAS using ensemble based positive-unlabeled learning 96%
- Exploiting Family History in Aggregation Unit-based Genetic Association Tests 94%
- An expanded analysis framework for multivariate GWAS connects inflammatory biomarkers to functional variants and disease 92%
Similar papers in this journal
- Controlling for background genetic effects using polygenic scores improves the power of genome-wide association studies 94%
- Network and pathway expansion of genetic disease associations identifies successful drug targets 94%
- Construction and optimization of multi-platform precision pathways for precision medicine 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.