The 2014 MAGES Conference hosted in Philadelphia brought out the shining stars in Statistical Genetics, along with a variety of approaches and difficulties researchers in the field are facing. Being my first MAGES event, I did not know what to expect; however, I was thoroughly impressed and am excited to go back next year.
Some of the topics that seemed to become more prevalent over the day included the use of BioBank data, algorithms taking advantage of Bayesian statistics, and talks addressing how complex disease actually is!
Dr. Dana Crawford of Vanderbilt University advocated utilizing BioBank data for research, including eMERGE and BioVU. BioBank data can be effective for researchers to use because they are inexpensive, provide large sample sizes and contain clinical data which they may not have had access to before. However, Dr. Crawford cautioned that there was the inability to follow-up on clinical data. Also researchers had to be careful since documentation has the potential to be inconsistent or spotty.
Several presenters incorporated Bayesian statistical methods including addressing eQTL studies and association testing. Dr. Joel Bader of Johns Hopkins University presented a study using “GWiS” meaning the Genome Wide Significance. Dr. Bader’s research focuses on identifying causal variants using a Bayesian heuristic approach to find the best fit without having to search every option.
The complexity of disease is not a new concept, in fact it is one of the most widely studied topics. Yet presenters at MAGES demonstrated that they have made great progress, but that there was still a long way to go with multiple ways of pursuing their goals. Dr. Marylyn Ritchie of Pennsylvania State University has had her approach dubbed as “PheWAS,” which aims to identify pleiotropic relationships and describe the genetic structure for complex traits.
Overall, the most interesting thing that I took away from the conference was the industry-wide call for collaboration. The goal has now shifted to networks of groups working together from across the United States, collaborating and combining data from all areas which will create a broader less biased view, and will in turn accelerate our quest for knowledge.