With the costs of genotyping and sequencing dropping rapidly, bringing together
individuals to collect phenotypes becomes a proportionally greater hurdle in genetic
research. As the flurry of genetic progress in the last decade has created an excitement
about genetics in the public, 23andMe has brought together over 150,000 consumers in
order to learn about their genetics and to participate in new research using web-based
surveys. This has allowed us to study over 1,000 different diseases and traits, most of
which have never been analyzed in large scale human genetics. I'll show how this approach
can lead to interesting results for a wide range of phenotypes, including Parkinson's
disease, allergies, and a wide variety of morphological traits. We have discovered over 250
novel associations for morphological traits such as shoe size, dimples, and nose shape.
These associations show that there is an interesting, complex genetic architecture
underlying human morphology and point towards a possible role of morphology driving
selection. I'll also discuss the current state of risk prediction based on genetics and give
some projections for the future.