This K18 application is to provide advanced mentorship in complex dynamic system analyses of clinical and biomarker data for major depressive disorder (MDD) and bipolar disorder (BD). MDD is a serious, debilitating illness that is projected to become the second global leading cause of disability by 2020. Bipolar disorder (BD) is a lifelong, chronic and highly recurrent, mood disorder characterized by episodes of mania or hypomania and episodes of major depression and is one of the top 10 causes of disability worldwide. The gap is that clinically relevant phenotypes and pathophysiology of MDD and BD are poorly delineated. One obstacle has been the prevalent use of static linear statistical measures to assess complex dynamic relationships between waxing and waning variables. Conventional analyses assume that signals are linear and stationary, and operate on a single (characteristic time) scale criteria that are routinely violated by real-world signals, i.e, they are highly nonlinear, operate on multiple time scales and are also nonstationary due to physiologic and pathologic phase transitions and other abrupt changes (bifurcations). The developmental aim of this K18 application is to learn methods to analyze complex dynamic nonlinear data to apply to the study of mood disorders with 12 months of intensive mentorship to the applicant through didactics and supervised analyses mentored by Ary L. Goldberger, Director of the Rey Institute for Nonlinear Dynamics in Medicine at Beth Israel Deaconess Medical Center and Wyss Institute for Biological Inspired Engineering at Harvard. The pilot project will analyze measures of arousal, heart rate, respiration, movement, and sleep stages using a 24 hour wearable devices from 5 patients with MDD, 5 age and sex matched patients with BP, and 5 matched healthy controls. These analyses will a) explore differences between MDD and healthy controls;b) BD and healthy controls;and c) MDD and BD. Methods for analyzing complex dynamic systems can open innovative new windows into understanding mood disorders. The applicant's long-term goal is to be able to apply nonlinear dynamics to the daunting problem of finding biologically relevant phenotypes and clinically useful biomarkers.
This research training program will teach the applicant how to use advanced statistical methods to analyze complicated data that vary over time from patients with mood disorders. Voice patterns, heart rate, breathing patterns, and movements over 24 hours are a few examples. These methods can lead to better diagnoses, better ways to understand the biology of mood disorders, and ways to measure response to treatment.
