Chapter 2 (Farrow et al. 2022; Brain) Supplementary tables

This repository contains the supplementary tables for chapter 2: Establishing gene regulatory networks from Parkinson's disease risk loci. See below for supplementary table descriptions.

Supplementary table 2.1: Nalls et al. 2019 GWAS dataset. Data downloaded directly from the publication Supplementary files. Odds ratios obtained from the IPDGC PD locus web browser.

Supplementary table 2.2: The Hi-C datasets used in CoDeS3D analysis. List of all Hi-C datasets used in this study and their reference to original article and GEO accession number

Supplementary table 2.3a: Nalls et al. 2019 GWAS: Sig eQTLs CoDeS3D output. Detailed data of eQTL-gene interactions when interrogating the 90 PD-SNPs. Overall 76 of the 90 SNPs tested were identified as eQTLs that were associated with the regulation of 518 genes across 49 tissues.

Supplementary table 2.3b: Replication of CoDeS3D whole-blood cis- eQTLs in eQTLGen

Supplementary table 2.3c: eQTLGen PD curated SNPs

Supplementary table 2.4: Nalls et al. 2019 GWAS: Sig eQTLs CoDeS3D output: Brain Hi-C: Brain GTEx only. Detailed data of eQTL-gene interactions when interrogating the 90 PD-SNPs in the 13 GTEx brain tissues only, and using brain-specific Hi-C cell lines (cortical plate neurons; germinal zone neurons; astrocyte of the cerebellum; brain vascular pericyte; brain microvascular endothelial primary cell; SK-N-MC; SK-N-DZ; neuronal progenitor cells; astrocyte of the spinal cord; dorsolateral prefrontal cortex cells; hippocampus cells; H1 neuronal progenitor cells)

Supplementary table 2.5: SNPnexus epigenome roadmap for all 90 SNPs, including histone modifications and DNaseI marker of open chromatin

Supplementary table 2.6a: Pathway analysis of the 518 genes; gProfiler (g:OST)

Supplementary table 2.6b: Pathway analysis of the 165 brain-specific genes; gProfiler (g:OST)

Supplementary table 2.7: The LOEUF values (downloaded from gnomAD) for each of the 518 genes included in the gene network. LOEUF score ("loss-of-function observed/expected upper bound fraction”) indicates the tolerance of a given gene to inactivation. A low LOEUF score indicates strong selection against loss-of-function variation.

Supplementary table 2.8: Nalls et al. 2019 GWAS: Sig eQTLs CoDeS3D output. Detailed data of eQTL-gene interactions when interrogating the 90 PD-SNPs in the foetal cortex, using 2 foetal Hi-C cell lines (cortical plate neurons; germinal zone neurons) and a foetal cortex eQTL dataset.

Supplementary table 2.9a: Gene regulatory network analysis. Protein:protein interaction analysis of 523 genes (using STRING, high confidence levels >0.700) followed by Louvain clustering identified 9 significant clusters.

Supplementary table 2.9b: Risk vs. protective proportion analysis of nine clusters

Supplementary table 2.10a: Pathway analysis for the genes within each of the nine significant clusters; gProfiler (g:OST) results (inc. bootstrapping analysis results)

Supplementary table 2.10b: Association between cluster enriched pathways and Parkinson's disease - reference table

Supplementary table 2.11: Sig eQTLs CoDeS3D output for rs10835060 and rs4238361: These two SNPs were identified by Makarious et al. as part of a polygenic risk score for PD diagnosis

Supplementary table 2.12: Correlational analysis – tissues falling outside of the 95% CI of expected eQTL numbers. Detailed data of the tissues that fall above or below the 95% CI of expected eQTL numbers, supplemental to figure 2. The data includes information for the analysis run on all 49 tissues (i.e. figure 2e-h).