Question: Protein-Protein interaction (Cytoscape)
0
gravatar for sindhujarajavelu
5 weeks ago by
sindhujarajavelu0 wrote:

Hi I am new to Cytoscape. I have constructed a network from human protein interactome using Cytoscape. I want to be sure that the network created wasn't by random. Can anyone suggest a way to validate it?

ADD COMMENTlink modified 4 weeks ago by xanderpico290 • written 5 weeks ago by sindhujarajavelu0

A random network is a network sampled from a random process or a probability distribution (on the edges/entries of the adjacency matrix). There are many models of random networks depending on the chosen generative process/probability distribution. So you need to define what you mean by random network.

ADD REPLYlink written 5 weeks ago by Jean-Karim Heriche23k

Thanks for the reply I want to make sure that the network created wasn't by random. I mean that it wasn't created by chance and want to prove that it represents a true biological network. Is there any way to validate?

ADD REPLYlink written 4 weeks ago by sindhujarajavelu0
1

I don't think you're asking the right question. The network you got is what it is by construction. You can characterize it and make hypotheses about the generative process, i.e. ask whether the network is compatible with a given random model. For this, check the networkrandomizer app and paper xanderpico mentioned in his answer. Keep in mind though that finding that a network is compatible or not with a particular random model doesn't tell you much about the biology underlying it. However you built it, your network is likely to be a poor representation of the biology, with biases, missing information ... but this is the information you have. The question should be what you can do with it that will give you insight into the biological question you're interested in.

ADD REPLYlink written 4 weeks ago by Jean-Karim Heriche23k
2
gravatar for xanderpico
4 weeks ago by
xanderpico290
xanderpico290 wrote:

There is an app for that: http://apps.cytoscape.org/apps/networkrandomizer. The associated paper may be helpful as well: https://f1000research.com/articles/5-2524/v3

ADD COMMENTlink written 4 weeks ago by xanderpico290
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1734 users visited in the last hour