Federated analysis for polygenic risk score calculations

Polygenic risk scores (PRS) provide an estimate of an individual’s disposition to a trait or complex disease which are calculated as the sum of the risk alleles weighted by the effect size estimate of the genome-wide association study data on the phenotype. PRS scores help in understanding the shared aetiology  of certain traits and also in risk prediction and prevention of certain diseases.   

We have been working on the development of a demonstrator of federated genetic analyses utilising a computational pipeline for PRS analysis. We have implemented this workflow using Nextflow, a modular and reproducible workflow manager that can be deployed and autoscaled to many different working environments including Slurm-based clusters and Kubernetes deployed using AWS amongst other computing environments. In this webinar, we will provide an overview of this PRS pipeline utilising the CINECA UK1 synthetic dataset, derived from the 1000 genomes project, as a demonstrator.We will further extend the demonstrator to other datasets and GWAS workflows, and integrate it to be run in a federated setting utilising the GA4GH guidelines on using Beacons for data discovery along with AAI passports for data access authorisation and deployments.   

The CINECA (Common Infrastructure for National Cohorts in Europe, Canada, and Africa) project aims to develop a federated cloud enabled infrastructure to make population scale genomic and biomolecular data accessible across international borders, to accelerate research, and improve the health of individuals across continents.

Computational biologists or similar interested in federated analysis and polygenic risk scores.

Explain federated data analyses for polygenic risk scores (PRS).

Describe the computational pipeline created for PRS analysis.