PaintorPipe

PaintorPipe performs Bayesian fine-mapping of GWAS loci by integrating GWAS summary statistics, linkage disequilibrium (LD) data, and functional annotations to prioritize causal single nucleotide polymorphisms (SNPs).

Key Features:

• Integration of Multiple Data Sources: Utilizes GWAS summary statistics, LD data, and functional annotations to rank genetic variants based on their likelihood of involvement in disease development.
• Bayesian Fine-Mapping Framework: Employs the PAINTOR framework to calculate the posterior probability of each SNP being causal using Bayesian fine-mapping.
• Functional Annotation Utilization: Incorporates functional annotations into the analysis to inform variant posterior probabilities and provide biological context for credible sets.
• Modularity and Customization: Modular design allows integration of diverse functional annotation sets for customized analyses.
• Visualization Capabilities: Produces visual annotations of credible sets using CANVIS.
• Technical Implementation: Implemented with Nextflow, supports execution locally or on SLURM clusters, and uses Singularity containers for reproducibility and portability.

Scientific Applications:

• Causal Variant Prioritization: Refines identification of putative causal variants from GWAS summary statistics by computing SNP-level posterior probabilities.
• Molecular Mechanism Inference: Links prioritized variants to functional annotations to support hypotheses about disease molecular mechanisms.
• Therapeutic Target Discovery: Highlights functionally annotated causal variants to aid identification of candidate therapeutic targets.
• Functional Interpretation of GWAS Loci: Integrates annotations to connect GWAS signals with biological pathways and regulatory features.

Methodology:

Uses GWAS summary statistics, LD data, and functional annotations as inputs; applies the PAINTOR Bayesian framework to compute SNP posterior probabilities; generates visualizations with CANVIS; implemented in Nextflow with support for SLURM execution and Singularity containerization.