Statistical methods for gene/protein expression and palliative care

点击翻译按钮获取中文摘要

My research proposal involves projects that are grounded in joint research with biological and health scientists.      Correspondence analysis (CA) is a descriptive technique designed to investigate the association between row and column variables by graphically displaying the patterns in the data.  I propose to explore and develop variations of CA techniques to identify differentially expressed genes and to investigate the use of CA to assess the quality of replicate DNA arrays.   Multiple correspondence analysis (MCA) and a related technique called joint correspondence analysis (JCA) are methods for visualizing the joint features of 2 or more categorical variables. We have been working with the Genetic Pathology Evaluation Centre (GPEC) at UBC and the Breast Cancer Outcomes Unit (BCOU) at the B.C. Cancer Agency to study relationships between molecular markers and outcomes for breast cancer.  Some researchers have incorporated survival information in an MCA analysis using supplementary points without regard to censoring.  I propose to develop techniques that accommodate possibly censored outcomes in MCA and JCA analyses of marker data.     For patients with life-limited illness, knowing how much time remains is often important for decision-making regarding goals of care, treatment options and dealing with closure on personal and family matters.  Research has shown that clinicians tend to overestimate survival times thus demonstrating the need for objective prognostic indicators.  The Palliative Performance Scale (PPS) is one such measure that, together with gender, disease and age, has been shown to be an effective predictor of survival.  Since the PPS trajectory cannot be treated as an external time-dependent covariate, I propose to develop joint longitudinal/survival models that account for the categorical nature of the longitudinal data.