This project is under development.

Note that it would require a significant investment of time, too much for the remainder of the course.

For those who are interested in following up with a semester-long capstone experience, this would be a good project.

To start out, the initial algorithm is called STRUCTURE, and is available from this link: http://pritchardlab.stanford.edu/structure.html.

The ADMIXTURE method is an extension of STRUCTURE. It uses the same statistical model, but uses a fast optimization. https://www.genetics.ucla.edu/software/admixture/

There are a couple of great resources for genetic data online.

• The first dataset is from the HapMap project. They have several reference populations available, as well as some admixed populations (i.e. - African Americans). You can go directly to the download page using this link (the readme has some of the details): ftp://ftp.ncbi.nlm.nih.gov/hapmap/genotypes/hapmap3/plink_format/draft_2/
• Also, the 1000s genomes project has a lot of data available as well: ftp://ftp-trace.ncbi.nih.gov/1000genomes/ftp/technical/working/20120131_omni_genotypes_and_intensities/