Skip to main content

Table 1 Betti number analysis and its implication for single protein targeting. The Surveillance, Epidemiology and End Results Database (SEER) provide data for the 5-year survival, degree-entropy was calculated from the Protein-protein interaction network (PPI) using Python®, Cycle-Basis was calculated using networkX.cycle_basis (a Python function), the nominal Betti number was calculated using JPLEX, and represents the Betti number for the full cancer network, and the Best Betti represents the lowest Betti number achieved by removal of a protein from the specific cancer PPI network. Degree-entropy and cycle basis are provided only as alternative measures of network complexity. When the elimination of proteins generated the same reduction in Betti, they were considered “equivalent targets”. A node connecting two clusters of proteins would be considered a node with high betweeness. We used highest betweeneness-centrality to differentiate equivalency targets, and the shaded entries have lowest Betti numbers and high betweeness centrality in the specific cancer

From: Design principles for cancer therapy guided by changes in complexity of protein-protein interaction networks

   AML Bladder CML Colorectal Endometrial Glioma NSCL Pancreatic Renal SCL Thyroid
5 years Survival [%] 23.6 78.1 55.2 63.6 68.6 33.4 18 5.5 69.5 6.2 97.2
Degree-Entropy 2.16 1.52 2.11 1.63 1.6 2.22 2.23 2 1.59 2.06 1.38
Cycle-Basis 108 20 115 58 45 128 75 72 52 150 24
Nominal Betti 107 20 114 51 45 128 50 38 51 149 24
Best Betti 95 15 101 41 35 109 37 29 34 131 17
Equivalent Targets FLT3          ITGA3  
HRAS MAPK3 AKT1 AKT3   HRAS     ITGA6 HRAS
NRAS MAPK1 AKT2 AKT2 PDPK1 NRAS KRAS KRAS HIF1A ITGA2B NRAS
KRAS   AKT3 AKT1 ILK KRAS    EPAS1 ITGB1 KRAS
          ITGA2  
          ITGAV