We split into two groups:

1. trying to run a GWAS, comparing the genotypes of openSNP users to the 1000 Genomes data set, to see whether there are significantly overrepresented variants in openSNP users
2. trying to recreate the main graphics of this publication, a graph of a principal component analysis which shows that genetic variation clusters well according to geography:

• Decided to only use data from 23andMe

• List and link your actual and ideal data sources.

• @gedankenstuecke
• @ciyer
• and other team members

• Repo for doing PCA on openSNP: https://github.com/ciyer/opensnp-fun