Hello there bloggies! Welcome to this beautiful fall Science Sunday!

First up, apologies for my incomplete posting last week! I was attending the American Society of Human Genetics (ASHG) annual meeting. I was overwhelmed by all of the science, meetings, and activities that I was unable to get my Trainee Tuesday and Fitness Friday posts out.

The meeting itself was filled with mostly great science! Though, they were a few cases of suspect science and press releases. Here’s a link to Ed Yong’s article over at The Atlantic discussing one of this suspect presentations about linking genetic modifications to sexual orientation (‘No, Sciences Have Not Found the Gay Gene’). Stay tuned to his writing in the next coming weeks for some of the better science that came from the meeting!

As always, going to the ASHG is a great way to gauge where the field is and where it may be going in the future. One thing I noticed was that the use of massive health-care based biobanks was definitely on the rise. The Nordic countries have long biobank traditions, but now Estonia and the United Kingdom are getting into the action. Even the United States is getting into the action with the forthcoming Precision Medicine Initiative. These collections bring together large amounts of electronic health records, other health data, and genetic information. One way researchers can take advantage of massive, centralized collections of health-related and genetic data is by performing phenome-wide association studies (Phe-WAS).

But what the heck is Phe-WAS?

Just like its name suggests, a Phe-WAS is similar to a GWAS or a genome-wide association study. Check out this previous post on what a GWAS is.

However, instead of scanning the every piece of genetic data for association with one trait or disease, researchers in a Phe-WAS take one genetic marker and see whether it is association with every disease or trait possible. For example, I could take a marker associated with Type II Diabetes and see whether it is also associated with neurological, cardiovascular, cancer, and other traits. It is a way to see whether genetic markers contribute to multiple traits and to investigate the underlying biological mechanisms of disease.

Having biobanks tied to electronic health records circumvents the common hurdle to PheWAS: access to lots and lots of clinical health data. Not only do they have lots of data, but they have loads of longitudinal data collected over many years. The development of disease can be monitored and then related back to genetics.

However, these data are collected for the clinic and not specifically for research. Many of these traits are for billing purposes, and misclassification is rampant. Plus, getting these data into a format that is usable in research is daunting to say the least. Complementing these health record data with addition data collected specifically for research would augment the amazing potential of these resources.

Thank you for stopping by Science Sunday! Would you allow your data to be part of a biobank? What do you think of the PheWAS approach? Let me know in the comments below or on twitter @DrFsThoughts.

See you later folks!

-Dr. F