Rebecca Betensky

Professor of Biostatistics

Department of Biostatistics

Research

Dr. Rebecca Betensky’s current methodological research interests are in the areas of latent class modeling for genomic data and survival analysis under complex sampling and with auxiliary information. Latent class models are useful for both unsupervised clustering of moderate to high dimensional genomic data and for supervised clustering by clinical outcomes, such as survival.  Dr. Betensky’s research involves the use of penalization, either in a frequentist or Bayesian setting, to enable model fitting with the high dimensional data.  Survival analysis under complex sampling arises in the context of cohort studies that undertake mid-study sampling for genetic analysis and that subset the data for focused disease related questions.  When the survival outcome is derived from a nonabsorbing process, there is often auxiliary information available than can be used to inform the survival analysis.  This research is motivated by problems that Dr. Betensky encounters in her collaoborations in neuro-oncology and multiple sclerosis and other neurologic diseases.