# PTL-PRS: an R package for transfer learning of polygenic risk scores with pseudovalidation

## TL;DR

PTL-PRS is an improved R package that enhances the accuracy of polygenic risk scores for underrepresented ancestry groups without needing individual-level data.

## Contribution

PTL-PRS introduces pseudovalidation and computational optimizations to improve usability and efficiency of transfer learning for polygenic risk scores.

## Key findings

- PTL-PRS eliminates the need for individual-level data by using pseudovalidation and pseudo-R2 metrics.
- Software optimizations in C++ and blockwise early stopping improve computational efficiency.
- PTL-PRS maintains predictive performance while being more accessible and privacy-preserving.

## Abstract

Polygenic risk scores (PRSs) are essential tools for predicting individual phenotypic risk but often lack accuracy in non-European ancestry groups. Transfer Learning for Polygenic Risk Scores (TL-PRS) addresses this challenge by leveraging European PRSs to improve prediction in underrepresented ancestries but requires privacy-sensitive individual-level data and has low computational efficiency. Therefore, we introduce Pseudovalidated Transfer Learning for PRS (PTL-PRS), an extension of TL-PRS that incorporates pseudovalidation to eliminate the need for individual-level data and includes further software optimization. For pseudovalidation, PTL-PRS generates pseudo-summary statistics for training and validation and evaluates model performance with the pseudo-R2 metric. To improve computational efficiency, PTL-PRS software was optimized with C++, blockwise early stopping, and direct genotype retrieval. Overall, PTL-PRS enhances usability while maintaining TL-PRS’s predictive performance.

The PTL.PRS R package is publicly available on GitHub at https://github.com/bokeumcho/PTL.PRS. The summary statistics used in this paper are available in the public domain: UK Biobank (https://pheweb.org/UKB-TOPMed), PGS Catalog (https://www.pgscatalog.org), COVID-19 Host Genetics Initiative (https://www.covid19hg.org) and GenOMICC (https://genomicc.org/data).