Skip to main content
Fig. 5 | Genome Medicine

Fig. 5

From: Pan-cancer identification of clinically relevant genomic subtypes using outcome-weighted integrative clustering

Fig. 5

Integration of multiple data types enhances the identification of survival distinct subgroups. a–h Each panel has two plots: the plot on the left summarizes median cross-validated logrank statistic across k = 2 to 8 (number of clusters). Each line is a data type (see legend), and the black line represents the survClust run on integrating all 6 platforms. Plot on the right summarizes the adjusted Rand index between cross-validated survClust solutions of individual data types and the integration of all. In this comparison, the survClust solution was chosen for an appropriate k which maximized logrank statistic and minimized the standardized pooled within sum of squares

Back to article page