Hi, I am working on protein interaction network using cytoscape. Once I prepare the network, and use some Network Visualizer ( like Cytoscape), I can get some network features. I wanted to know about the significance of these network values like node degree and centrality measures such as Betweenness Centrality, Closeness Centrality, etc. Also, How to decide the hub proteins with these values
Hi Rajasekar,
First, to understand whether a node is hub or not you need to calculate the parameters for individual nodes. The parameters calculated for the entire network reflects overall concept but wouldn't help in such cases.
Degree shows the number of connected nodes with the individual node. So higher degree indicates a characteristics of hub.
Similarly, the more central (closeness centrality) a node is, the closer it is to all other nodes. So higher closeness centrality reflects the tendency of a node to be a hub.
Regards,
Arindam.
Hi Rajasekar
Just to add up to your answer. I understand you want to explore the interactions of the PPI. But coming to your final question on hub, let me start by saying that different researcher may have slightly different opinion on what is a hub. For instance, if the average degree of every node (or vertex) in the network is 3, one may say a hub is a node with more than 3 nodes or even stricter, hub may be node with more than 5 nodes. Hence, hub will depend on the context or your area of work. But generally a hub will have greater degree of nodes than the average node degree in network.
In your case to find out the degree you may write a java program and use Cytoscape API to find out the degree or you may also directly write a small program (java, c, c++, etc ) to count the degree, centrality , betweeness etc
You can get some idea about the algorithm in case you want to try out:
Article Weighted Edge Based Clustering to Identify Protein Complexes...
Article Construction and Analysis of the Protein-Protein Interaction...
Duplicate of this https://www.biostars.org/p/280679/
In PPI networks those hubs having more nodes are considered to be relevant. In order to evaluate the parameters that determines the hub integrity, a Cytoscape plugin, cytohubba is freely available. Kindly go through the below link for more details of cytohubba:
http://hub.iis.sinica.edu.tw/cytohubba/
For more details oh hub identification please go through this article:
Article Identifying Hubs in Protein Interaction Networks
I hope this helps you.
Best,
M
Hi Rajasekar,
I'm not quite sure if it's relevant to what you're working on or not. But anyway, if you made any network with the purpose of finding the right candidates, you may consider using the IVI, an integrative algorithm, for finding the most influential nodes within the network! It is included as a function in the Influential R package (https://cran.r-project.org/web/packages/influential/index.html).
You may also read the corresponding paper at https://www.researchgate.net/publication/342374701_Integrated_Value_of_Influence_An_Integrative_Method_for_the_Identification_of_the_Most_Influential_Nodes_within_Networks
Use cytohubba plugin cytoscape which have 11 clusetering methods.
to find module on the basis of centrality pls see the link below
Article Identification and Validation of Potential miRNAs, as Biomar...
https://www.mdpi.com/search?q=&authors=mansoor+ali&journal=genes&article_type=&search=Search§ion=&special_issue=&volume=&issue=&number=&page=
- Badges
- Science topic
- Similar topics
- Comment