It can arise that underused data analytic methodology may enjoy wide-ranging applicability. I showcase use of the Patient Rule Induction Method (PRIM) at two radically different scales: (1) identifying genomic '3D hotspots' -- localized regions wherein an attribute superposed on a 3D genome reconstruction is extreme; and (2) identifying stellar kinematic groups and open clusters in the Milky Way. In the first example a novel principal curve based reconstruction technique is introduced while, in both instances, nuances surrounding coordinate systems and inference are addressed.
Speaker: Mark Segal, PhD, Professor, Department of Epidemiology and Biostatistics, University of California, San Francisco. Segal focuses on the development and application of statistical methods to address problems in computational biology and genomics. He has devised methods for addressing several aspects of analyzing data deriving from high-throughput biotechnologies, straddling low-level (e.g., pre-processing) to high-level (e.g., linked survival phenotypes, regulatory module elicitation) approaches.
Social hour following: 4-5PM