Customer:

The Feinstein Institute for Medical Research is a unique, disease-oriented biomedical research institution dedicated to advancing science, shaping medicine and making a difference in people’s lives.  More than 100 doctors and scientists study disease at all levels — from molecular biology to patient-oriented research — and translate their scientific findings into practical use. Located in Manhasset, NY, The Feinstein Institute is one of the fastest-growing biomedical research institutes in the United States.

Problem:

Among the numerous studies under way at the Institute, Dr. Todd Lencz is focused on schizophrenia, specifically trying to find the genetic causes of this disease that plagues approximately 1% of human adults.  To put this in perspective, in the US there are an estimated 2.4M people suffering from schizophrenia at any given time.  Further, fifty percent of people in hospital psychiatric care have been diagnosed with the disease.  The real costs of this disease are staggering, and the human suffering caused is inestimable. 

Once a genetic link is found, drugs or treatment can be identified or designed to offset the effects caused by the genetic mutation that ultimately causes the disease.

Methodology:

Finding genetic links to disease is no small proposition.  There is tremendous cost and effort required to simply collect DNA samples from enough people to create a statistically valid sample.  Turning those samples into data that can be analyzed requires sophisticated equipment and personnel.  Yet when all is said and done, there is still the need to analyze a massive amount of data, looking for the proverbial needle in a very large haystack. 

In the case of this study, the data had 500,000 categorical genetic variables for each of the several thousand patients.

The first steps in the analysis were data cleanup and quality using HelixTree (a version of Optimus RP customized for genetics).  With no further augmentation or manipulation, the entire dataset was imported into HelixTree, and an exhaustive association study was performed.

Results:

The study was the first successful whole genome association study published for a psychiatric illness.  A significant medical research paper was published (see below) on an important medical finding that may lead to treatment.