Genome‐wide Association and Polygenic Risk Score Analysis
After the human reference genome release, genome wide association studies (GWAS) have been performed to quantify the statistical association between common genetic variation and disease outcomes or quantitative traits, risk factors for diseases. Thousands of single nucleotide polymorphisms (SNPs) associated with hundreds of risk factors and diseases have been published, but the biological mechanisms underlying most of the diseases are still unknown.
Are the identified variants actually causing the disease? What is the impact of the environment vs genetics in risk for diseases? What is the biological function of the variants mapping to the non-coding regions of the genome? Is it possible to predict the individual population-wide disease risk?
These are some of the challenges of the Post-GWAS Era that we will discuss during this two-day workshop. The sessions will alternate foundational concepts of genomics and population genetics and practical analyses using genetic tools to perform genome-wide linear and logistic models, and to visualize the results using a Manhattan and quantile-quantile plots using online resources and R.
This is an introductory workshop in the Computational Genomics series. No prior experience is required. Understanding of simple statistical and genetics concepts and bash/R basics is required.
Day1: Genetic concepts and demo I: genotypes QC, population stratification and visualization.
Day2: Methodology concepts and demo II: GWAS for qt and binary traits, visualization and PRS estimation.
The slides are here.
There is a zip file with code and data at this link.
- Open the link, press the download button (usually on the top right corner), and select “Direct Download” if a menu pops up.
- The zip file download should start.
- Unzip the zip folder (handson_plink ~900 Mb).
- Please move the handson_plink/ folder to Desktop or to a preferred location
The folder include scripts and input files the we will be using in both session:
Day 1: step01 and step02
Day2: step03 through step06
The demo will use PLINK2 command line and simple R scripts.
Please note that the hands-ons part has only been tested on macs or linux machines.
Please install PLINK2 on your laptop
- Download the version based on your OS from 'binary downloads' session - column 'Development (21 Nov)'
- Double click on the zip folder
- Right-click on the unzipped executable file 'plink2' - if asked about opening an app from an unidentified developer, please click 'open'.
- Please copy plink2 executable file to ~/Desktop/handson_plink/ (or preferred location)
- Open your terminal and type ./plink2 to verify the correct installation
- Open Rstudio
- Install the package required:
install.packages("qqman")
- Verify the qqman installation
library("qqman")
No error/warning message returned