pathwayPRS_paper

Code used in the manuscript investigating associations between pathway-specific PRS, Alzheimer's disease diagnosis and ATN biomarkers

Authors

• Nicholas J. Schork - The Translational Genomics Research Institute
• Jeremy A. Elman - University of California San Diego

Citation

Schork, N. J., Elman, J. A., & Alzheimer's Disease Neuroimaging, I. (2023). Pathway-Specific Polygenic Risk Scores Correlate with Clinical Status and Alzheimer's Disease-Related Biomarkers. J Alzheimers Dis, 95(3), 915-929. doi:10.3233/JAD-230548. Link to article

Prerequisites:

1. MAGMA
2. PRSice
3. Cytoscape
4. EnrichmentMap
5. AutoAnnotate
6. ADNIMERGE
7. R
8. PLINK

Data requirements

1. Summary statistics from the Kunkle et al. 2019 GWAS of AD (PMID: 30820047) are available from NIAGADS
2. GO (no iea), KEGG, and Reactome gene sets from Bader Lab (February 2, 2022)
3. Download the following files from the MAGMA webite:
   • Gene locations, build 37
   • SNP synonyms, dbSNP 151
   • 1000 Genomes EUR reference data
4. Gene boundaries from GENCODE
5. ADNIMERGE R Package from the ADNI website.

Code

1. create_genesets.sh - Clean, format, and concatenate gene sets from Bader Lab
2. run_magma.sh - Run MAGMA gene analysis on AD GWAS summary statistics
3. magma2enrichmentMap.R - Convert MAGMA output to EnrichmentMap input
4. createPathwayClusters.R - Create pathway clusters using Cytoscape apps EnrichmentMap and AutoAnnotate
5. cytoscape2gmt.R - Convert Cytoscape cluster output to GMT file for use in PRSice
6. run_PRSice.sh - Calculate pathway-specific PRS for ADNI participants using PRSice and AD GWAS summary statistics
7. prsAssociation.R - Run linear regression models to test associations between pathway-specific PRS and ATN biomarkers