New implementation of Structure is orders of magnitude faster for large data sets. The data they simulate in the paper seems very much inline with the size of our own data sets (600 individuals x 2500 loci).

This should make it easier to apply to our genotyping-by-sequencing data sets.

Raj, Anil, Matthew Stephens, and Jonathan K. Pritchard. “Variational Inference of Population Structure in Large SNP Datasets.” Genetics (2014): genetics-114.

See also:
Novembre, J., Variations on a Common STRUCTURE: New Algorithms for a Valuable Model,
Genetics July 2014 197:809-811; doi:10.1534/genetics.114.166264