Table 1 Average human interactome centralities of proteins related to colorectal cancer and type 2 diabetes.

From: Targets of drugs are generally and targets of drugs having side effects are specifically good spreaders of human interactome perturbations

Centrality type

Disease-related proteins

Proteins, which are not related to any of the two diseases

 

Colorectal cancer

Type 2 diabetes

Statistical difference between cancer- and diabetes-related proteins

Centrality value

Statistical difference from values of cancer-related proteins

Statistical difference from values of diabetes-related proteins

Degree (number of neighbours)

159.5

9.000

7.09e−5

9.000

2.58e−9

0.830

Closeness centrality (1/edge)

0.357

0.294

3.46e−5

0.277

1.90e−10

0.122

Betweenness centrality (fraction of shortest paths passing through the node)

2.55e-3

1.16e-5

1.24e−4

1.34e-5

3.23e−9

0.922

  1. The table shows the medians of the centralities of proteins related to colorectal cancer and type 2 diabetes (results were very similar, if instead of medians we used their arithmetic means; data not shown). The total number of colorectal cancer- and type 2 diabetes-related proteins was 18 and 14, respectively. Centrality values were calculated with the Pajek programme57. The human interactome containing 12,439 proteins and 174,666 edges was built from the STRING database46. Colorectal cancer-related proteins were obtained from the Cancer Gene Census database48, type 2 diabetes-related proteins were obtained from the article of Parchwani et al.49. Statistical analysis was performed using the Wilcoxon rank sum (Mann-Whitney) test function of the R package56.
Back to article page