R GWAS II: Install packages and software

Downloads

1. Go to https://www.cog-genomics.org/plink2 and download PLINK. It is an executable file so you don’t need to install it.
2. Get a copy of the dataset gwas-data.ped, gwas-data.map and covs.txt from the USB stick provided at the exercise. Alternatively, get the bed/bim/fam files, they are smaller and contain the same information but in binary format. You can use PLINK to recode the binary files to the ped/map format later using the --recode command.

The example data includes SNP data from 168 individuals and about 730.000 SNPs.

The file gwas-data.map contains information about the SNP locations. Map-files always have the same structure:

1. Chromosome in the first column
2. SNP name in the second column
3. Genetic distance (in centimorgan) in the third column
4. Base pair position in the fourth column.

In GWAS the genetic distance can be set to 0 for all SNPs. In this case it is actually provided.

The gwas-data.ped stores all the genotypes (this file is big so don’t try to open it). The columns are:

1. Family ID
2. Individual ID
3. Paternal ID
4. Maternal II
5. Sex (1=male, 2=Female)
6. Phenotype
7. Genotypes in the following columns (2 columns per SNP)

The file covs.txt stores two alternative phenotypes: a disease (categorical) trait c and a continuous (quantative) trait q. And some covariates. Open it in a text editor and see whats there.

To make the rest of the exercise a bit easier to work with you can copy the executable plink file into the folder where you store your data.

If you chose to copy the bed/bim/fam files from the USB you can get the ped/map files by running the command:

plink --bfile gwas-data --recode --out gwas-data

at the command line.

---

Next: Quality-control

Previous: Instructions