Phenotype definition of psychiatric disorders in genetic association studies

Andrew Liu is giving a talk on genetic association studies on Friday, March 2, 2012. The summary of the talk is as follows:

It is well known that genetic factors can play an important part in the incidence and development of many psychiatric disorders.  While demonstrating the role of genetics in these disorders has been relatively simple (twin studies, heritability studies), determining the exact location of the disease-related gene locus has proven difficult, with most discoveries failing to be replicated.  Two main reasons for this is the subjectivity involved in defining phenotypes, as well as the phenotypic heterogeneity of subjects who have received the same diagnosis.

In this talk, we give a brief overview of common methods used in genetic epidemiology, and discuss statistical methods used in genetic association studies.  We discuss ways of better defining psychiatric phenotypes, and apply these ideas to a pediatric Attention Deficit Hyperactivity Disorder (ADHD) dataset to obtain a valid, objective phenotype for response to medication in children with ADHD.  This phenotype is then compared to an alternate measure in the context of a family-based genetic association study.