Biostatistics, Computation and Data Management Core
Personnel:
Hal Stern
Chancellor’s Professor, Department of Statistics
David Keator
Reacher Professor, Department of Psychiatry and Human Behavior
Anton Palma
Principal Biostatistician, Institute for Clinical and Translational Science
Hina Arora
Graduate Student, Department of Statistics
Core Aims:
The projects of the proposed Conte Center integrate human and experimental animal studies to investigate the overarching hypothesis that fragmentation and unpredictability of early life maternal and environmental signals (FRAG) contribute to adolescent vulnerabilities and subsequent mental illness via mechanisms involving aberrant development and maturation of emotional brain circuits. The Biostatistics, Computation and Data Management (BCDM) Core works closely with Projects 1-4 and the Imaging Core to plan and support their statistical analyses, to develop innovative models that predict risk for mental illness, and to provide data management support. This work is achieved through addressing the following aims:
Aim 1: Provide computational and biostatistics support for studies of the relationship of FRAG with age- and sex-specific vulnerabilities and mental illness. Address the contributions of FRAG in the context of other risk factors and optimize approaches to the analysis of complex datasets that include cognitive, emotional, and imaging data.
Aim 2: Use data from multiple human cohorts to develop and test measures that easily, reliably, and accurately assess fragmented and unpredictable maternal and environmental signals.
Aim 3: Develop statistical / computational models that integrate measures of early-life experiences with cognitive, emotional and imaging data to predict anhedonia and risk for mental illness.
Aim 4: Develop a Center information resource, provide data management support to all projects and cores, and facilitate integration across projects. This includes accommodating human demographic, cognitive, emotional, and imaging data and experimental animal data.
Research:
The BCDM core has delineated a novel measure of patterns of behaviors across species. Specifically, the core computed Entropy rate as a measure of unpredictability.
Defining entropy for mother-child interactions
Precise methodology is in Davis et al., 2017 and on our resources page
Entropy can be measured across species (Left: human subjects; Right: rodents)
Entropy rate (1 yr of age) is associated with child cognitive performance.
