Title: Polygenic Scores in Practice: Standardisation, Missing Variants, and Lessons from GenoPred

Abstract: 

Polygenic scores (PGS) are increasingly used in research and clinical settings, but their reliability depends on how they are calculated and applied across diverse datasets. This talk will outline practical considerations for deriving and applying PGS in external cohorts, including standardisation, handling missing variants, and ensuring comparability across studies. I will draw on lessons from developing the GenoPred pipeline, which implements best practices for reproducible scoring and facilitates equitable prediction across populations. Examples will illustrate common pitfalls and solutions, and highlight ongoing challenges in making PGS robust and clinically meaningful.

Speaker Biography: 

Oliver Pain is a Sir Henry Wellcome Postdoctoral Fellow at King’s College London. His research focuses on developing and applying statistical genetics methods to improve the accuracy, equity, and clinical utility of polygenic scores. He leads the development of GenoPred, a reproducible pipeline for polygenic scoring, and collaborates widely across international consortia on genetic prediction of complex traits and treatment response.

Use this Teams link to join the call. 

Please contact maudsley.brc@kcl.ac.uk if you have any questions.