could not interpret terms "node centrality", "betweenness centrality" "suboptimal paths"

[Kajwal Kumar Patra]

Hello sir,

can you please simplify these above terms and also can you let us know how to implement these in network analysis.

Also in the tutorial, "cna_vintage.pdf(protein structure network with bio3d)" , page# 20 , figure:19, i have confusions on path length distributions as follows: 1. 1st figure shows the probability density axis starts from 0 to 120. Is that obvious?(i mean max. proabity can be 1, forgive me if i am wrong) 2.the path length in x axis is in angstrom(i hope).

3. you have defined node degeneracy as number of paths passing through each node; now in the Node degeneracy figure, it seems that node#32 has 50 number of path, so for me the node degeneracy of node#32 is 50 instead of 1(as the node degeneracy table shows). table also shows for node#46, the degeneracy is 0.12, what does that mean?

looking forward your response, kajwal

[Xinqiu Yao]

Hi,

All these terms and how they can be implemented with bio3d have been documented in the "cna_vignette.pdf". Please read the tutorial carefully and post more specific questions if you are still unclear after reading.

1. It is probability density not probability. Probability density can be larger than 1 but the integration over x-axis is 1.

2. Path length here indicates the coupling strength between two distal sites and does NOT represent the length of a "physical" path. Mathematically, path length is the sum of -log(cij), where cij is the correlation value between two residues along the path. Therefore, it does not have a unit.

3. The node degeneracy shown in the table is normalized. You can print non-normalized with print(pa, normalize=FALSE) and then you get the same results as shown in the figure.

[Kajwal Kumar Patra]

Dear sir,

in sub optimal path calculations,

pa <- cnapath(net, from=32, to=131, k=50

when k=50, we explore 50 suboptimal paths. But when i dont specify the k value, it explores 10 paths, so (i) Is 10 the default vaue of k? and (ii)how can i know the exact no of paths between the source and the sink?..

[Xinqiu Yao]

Yes, the default value of k is 10. There is no functionality to estimate the "exact" number of paths but usually we set 500 and it will give pretty sufficient statistics for the underlying problem.