PGSXplorer: an integrated nextflow pipeline for comprehensive quality control and polygenic score model development

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Next-generation sequencing and genomic data sharing have advanced individualized medicine.
• Genome-wide association studies (GWAS) are crucial for understanding complex diseases and developing polygenic scores (PGS).
• Limited standardized pipelines hinder the widespread adoption and generalizability of PGS.

Purpose of the Study:

• To introduce PGSXplorer, a comprehensive Nextflow DSL2 pipeline for automated PGS development.
• To address the need for standardized, scalable, and reproducible PGS construction.
• To improve the generalizability of PGS through multi-origin data integration.

Main Methods:

• Developed PGSXplorer as a Nextflow DSL2 pipeline using Docker encapsulation.
• Integrated quality control, phasing, imputation, and PGS model construction functionalities.
• Incorporated multiple PGS development tools (PLINK, PRSice-2, LD-Pred2, Lassosum2, MegaPRS, SBayesR-C, PRS-CSx, MUSSEL).

Main Results:

• PGSXplorer automates the end-to-end process of PGS development.
• The pipeline demonstrated scalability and effectiveness on synthetic datasets for single- and multi-population analyses.
• Integration of diverse tools enhances the generalizability of developed PGS.

Conclusions:

• PGSXplorer provides a standardized and automated solution for PGS development.
• The open-source tool facilitates advancements in precision medicine and genetic research.
• Availability with tutorials promotes wider adoption and reproducible research.