Semi-supervised recursively partitioned mixture models for identifying cancer subtypes

Abstract

Motivation: Patients with identical cancer diagnoses often progress differently. The disparity we see in disease progression and treatment response can be attributed to the idea that two histologically similar cancers may be completely different diseases on the molecular level. Methods for identifying cancer subtypes associated with patient survival have the capacity to be powerful instruments for understanding the biochemical processes that underlie disease progression as well as providing an initial step toward more personalized therapy for cancer patients. We propose a method called semi-supervised recursively partitioned mixture models (SS-RPMM) that utilizes array-based genetic and patient-level clinical data for finding cancer subtypes that are associated with patient survival.

Results: In the proposed SS-RPMM, cancer subtypes are identified using a selected subset of genes that are associated with survival time. Since survival information is used in the gene selection step, this method is semi-supervised. Unlike other semi-supervised clustering classification methods, SS-RPMM does not require specification of the number of cancer subtypes, which is often unknown. In a simulation study, our proposed method compared favorably with other competing semi-supervised methods, including: semi-supervised clustering and supervised principal components analysis. Furthermore, an analysis of mesothelioma cancer data using SS-RPMM, revealed at least two distinct methylation profiles that are informative for survival.

Availability: The analyses implemented in this article were carried out using R (http://www.r.project.org/).

Contact: devin_koestler@brown.edu; e_andres_houseman@brown.edu

Supplementary information: Supplementary data are available at Bioinformatics online.

Fig. 1.

Average pseudo-R² between SS-RPMM, SS-Clust, SPCA using the first principal component only SPCA(1), and using the first and second principal components SPCA(2), for different settings of M and K. Results are based on 100 simulations.

Fig. 2.

Heatmap of predicted class memberships for the observations in the testing set using the average beta values for the 41 loci with largest absolute Cox-scores. Observations within predicted class as well the 41 loci were clustered using hierarchical clustering with Ward linkage and Euclidean distance metric.