Figure 1

Analytical pipeline, clustering results and GO term distributions. A: Detected proteins in mod- and poor-stage HCC patients are used to build PDSs (Pathway-Derived Subnets) from an integrated pathway database (PathwayAPI). These PDSs are used for calculating hit rates for each patient to generate a PSP. The set of PSPs are used for sample class analysis as well as significant feature identification. B: Sample class analysis PDSs have sufficient resolution to segregate mod- and poor-stage patients with high confidence. C: Significant GO term distribution A large number of significant GO terms are associated with metabolic functions. Although cancer-associated terms such as apoptosis, growth and immune responses are also uncovered. This is consistent with earlier observations based on this dataset