Skip to content

songlab-cal/StratPGS

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

41 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Welcome to StratPGS

This repo contains results and scripts for reproducing and repurposing our analyses in our work,

Alan Aw, Jeremy McRae, Elior Rahmani and Yun Song (2024+) "Highly parameterized polygenic scores tend to overfit to population stratification via random effects" bioRXiv preprint. DOI: 10.1101/2024.01.27.577589

Dashboard

As the image above shows, we also present our results in an interactive dashboard, available here.

Directory Structure

Results

Under results:

  • random_projections contains results from Performance Inflation by rPGS
  • effect_perturbation contains results from Performance Relative to pPGS and sPGS
  • PGS_catalog contains results from analyses of polygenic risk scores obtained from the PGS Catalogue (Evaluation of MCH PGSs in our paper)
  • PGS_perf_vs_pval_thres contains the data file and plot showing the performance, averaged across traits, of PGSs we trained on UKB phenotypes using various GWAS p-value thresholds

Scripts

Under scripts, similar directory organization as Results. We include one additional subdirectory:

  • angular_central_gaussian, which contains a script to simulate random vectors under the Angular Central Gaussian distribution. This is mentioned briefly in our Main Text and discussed in our Supplementary Material

Logs

Under logs, we provide log files that record the statistical tests we performed, as described in our paper.

Manuscript

Under manuscript, we provide the Supplementary Material for our work, which contains technical details and more mathematical ideas that unfortunately could not fit into the Main Text.

Notes

  1. The summary of non-zero variant counts of all PGS Catalogue scores is available under results/PGS_catalog/all_polygenic_scores. The associated data wrangling script is available under scripts/PGS_catalog/all_polygenic_scores.

About

Performance Inflation and Perturbation Sensitivity of PGSs

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published