B0418
Title: Bayesian bi-clustering for temporally heterogeneous high-dimensional longitudinal data
Authors: Massimiliano Russo - Harvard Medical School (United States) [presenting]
Abstract: X-linked dystonia-Parkinsonism (XDP) is a rare genetic form of dystonia found almost entirely among males of Filipino descent, characterized by highly heterogeneous symptoms and unknown progression patterns. Distinguishing subtypes of XDP is pivotal in advancing the understanding of the disease and providing effective targeted treatment for the affected patients. However, analysis of existing data is complicated by the fact that (i) the patients are observed for only a short length of time at different stages of progression, and (ii) the disease symptoms are measured using a large number of interdependent scales. To overcome these challenges, a novel Bayesian statistical model is proposed that simultaneously clusters subjects according to the trajectory of their progression and clusters variables into jointly relevant aspects of the disease. The model is applied to clinical XDP data and is found to reveal novel insights into the patterns of disease progression.
