Abstract: Cellular signal transduction systems are organized as hierarchical network. When stimulated by environment changes, cellular signals are transmitted through signaling cascades in which signals are compositionally encoded. For example, the signal of an activated growth factor receptor, EGFR, is then compositional encoded by RAS, PI3K, AKT, and then by STAT3 and cJUN etc. Often, the effect of perturbation of cellular signaling system can be read out as changed gene expression.
Abstract: This talk will provide an overview of the new Center for Causal Discovery (CCD), which recently was funded as an NIH Big-Data-to-Knowledge (BD2K) Center of Excellence. The CCD is focused on developing and disseminating computational methods for causal modeling and discovery of biomedical knowledge from big data. Its aims include research, training, software dissemination, and collaborative projects with other BD2K Centers of Excellence.
Abstract: The nature of diagnostic healthcare is changing dramatically, thanks in part to the development of cost effective technologies for whole slide digital scanning, and high-throughput genetic, genomic and epigenetic data collection. There is a consensus among clinicians and researchers that integrative, co-analysis of clinical, genomic, histopathological and radiological data will open new and important venues for personalized medicine strategies in the near future.
Deriving predictive models in medicine typically relies on a population approach where a single model is developed from a dataset of individuals. We develop and evaluate a personalized approach in which we construct a new type of decision tree model called decision-path model that takes advantage of the particular features of a given person of interest. We show that the personalized approach of learning decision tree models i can perform better than a population approach for predicting clinical outcomes.