RICOPILI: Rapid Imputation for COnsortias PIpeLIne

Area of Science:

• Genetics and Genomics
• Computational Biology
• Statistical Genetics

Background:

• Genome-wide association studies (GWAS) are crucial for understanding complex traits.
• Processing large-scale, multi-cohort GWAS data presents significant computational challenges.
• Existing pipelines may lack efficiency or adaptability for diverse high-performance computing environments.

Purpose of the Study:

• To develop RICOPILI, an open-sourced Perl-based pipeline for efficient processing of large-scale GWAS data.
• To address challenges in quality control, imputation, and downstream analyses for multi-cohort GWAS.
• To provide a computationally efficient and portable solution for genetic research.

Main Methods:

• Developed RICOPILI, a Perl-based pipeline integrating automated parallelization and cluster job management.
• Implemented features for technical and genomic quality control (QC) in various cohort types.
• Included genome-wide phasing, imputation, association analysis, meta-analysis, polygenic risk scoring, and replication analysis.

Main Results:

• RICOPILI demonstrates computational efficiency and portability across diverse high-performance computing (HPC) environments.
• The pipeline successfully automates key GWAS steps, including rapid production of imputed genome-wide data.
• Integrated meta-analysis of simulated GWAS data validates each pipeline step with visualization plots.

Conclusions:

• RICOPILI provides a robust and flexible platform for large-scale GWAS, facilitating data interpretation and manuscript preparation.
• Its adaptable architecture supports ongoing development and integration of new algorithms.
• The pipeline is a valuable tool for the psychiatric genomics community and beyond, aiding genetic discovery.