Survival-Supervised Latent Dirichlet Allocation Models for Genomic Cancer Research

• Activity Number: 253
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): John A. Dawson*+ and Christina Kendziorski
• Companies: University of Wisconsin-Madison and University of Wisconsin-Madison
• Keywords: Survival analysis ; Topic models ; Latent Dirichlet allocation ; Cancer ; Genomics

Abstract

Two challenging problems in the clinical study of cancer are the characterization of cancer subtypes and the classification of individual patients according to those subtypes. Statistical approaches addressing these problems are hampered by population heterogeneity and challenges inherent in data integration across high-dimensional, diverse covariates. We have developed a survival-supervised latent Dirichlet allocation (LDA) modeling framework to address these concerns. Our framework extends LDA to the genome by considering each patient as a "document" with "text" constructed from clinical and high-dimensional genomic measurements. We then extend the traditional LDA framework to allow for topics whose generation is supervised by time-to-event responses. The model enables the efficient identification of collections of clinical and genomic features that co-occur within patient subgroups, and then characterizes each patient by those features. An application to The Cancer Genome Atlas (TCGA) ovarian project identifies patient subgroups that are characterized by different propensities for exhibiting abnormal mRNA expression and methylations, corresponding to differential survival rates.