Skip to content

Latest commit

 

History

History
85 lines (59 loc) · 3.77 KB

README.md

File metadata and controls

85 lines (59 loc) · 3.77 KB
Raw

🌟 postGWAS_Colocalization

We provided a step-by-step guide for performing colocalization analysis using the coloc package in R. Image text

📊 Colocalization Visualization Guide

This guide provides a detailed walkthrough on the visualization of colocalization results in R. The visualization aims to represent the association of specific SNPs with two different phenotypes, typically GWAS and eQTL datasets, to understand shared genetic architectures.

The script coloc_Visualizition.R provides a step-by-step approach to generate the colocalization plot.

🛜 Download and install coloc

if(!require("remotes"))
  install.packages("remotes")
  install.packages("dplyr")
library(remotes)
install_github("chr1swallace/coloc",build_vignettes=TRUE)
library("coloc")
library(dplyr)

📥 Importing Data

🧬 Import Phenotype 1 (GWAS) data

gwas_data <- read.table("path_to_gwas_data.txt", header=TRUE, sep="\t")
  • ⚠️ Replace path_to_gwas_data.txt with the path to the correct GWAS dataset.

🧪 Import Phenotype 2 (eQTL) data

eqtl_data <- read.table("path_to_eqtl_data.txt", header=TRUE, sep="\t")
  • ⚠️ Replace path_to_eqtl_data.txt with the path to the correct eQTL dataset.

🔄 Merging GWAS and eQTL data

Before performing the analysis, ensure that the datasets are merged based on the SNP IDs.

merged_data <- merge(gwas_data, eqtl_data, by="SNP_ID")

Replace SNP_ID with the appropriate column name for SNP identifiers if it's different in datasets.

📊 Performing Colocalization Analysis

With the datasets imported and merged, performing the colocalization analysis.

library(coloc)

results <- coloc.abf(dataset1=list(beta=merged_data$beta_GWAS, varbeta=merged_data$varbeta_GWAS, pvalues=merged_data$pvalue_GWAS, type="quant trait", N=sample_size_gwas),
                     dataset2=list(beta=merged_data$beta_eQTL, varbeta=merged_data$varbeta_eQTL, pvalues=merged_data$pvalue_eQTL, type="quant trait", N=sample_size_eqtl))
  • ⚠️ Columns "pvalue_GWAS" and "pvalue_eQTL" contain the p-values for GWAS and eQTL respectively
  • ⚠️ Replace sample_size_gwas and sample_size_eqtl with the actual sample sizes for the correct GWAS and eQTL studies, respectively.

🔍 Interpreting Results

The coloc.abf function outputs posterior probabilities for each of the five hypotheses (H0 to H4). In general, a high posterior probability for H4 suggests that the GWAS and eQTL signals colocalize, indicating that they are likely driven by the same causal variant.

📚 Resources for Colocalization Analysis

Below are some essential resources and datasets for performing colocalization analysis:

🌐 eQTLs from 37 datasets

This resource consolidates eQTL data from 37 different datasets, encompassing a total of 31,684 individuals. It provides a rich collection of eQTLs that can be utilized for various genomic studies including colocalization analysis.

eQTLgen datasets

📊 GTEx Datasets

The Genotype-Tissue Expression (GTEx) project provides a wide variety of datasets spanning multiple versions (from v6 to v8). These datasets are invaluable for studying the relationship between genetic variations and tissue-specific gene expression.

GTEx Portal datasets

📦 coloc R Package

The coloc package for R is an essential tool for performing colocalization analysis. It allows researchers to test for shared causal variants between two phenotypes, using summary data.

coloc GitHub Repository & Documentation