Questions and Answers

What are the Different Types of Networks?

There are two types of networks displayed on this site.  Both are two forms of gene co-expression networks:

  1. Global Networks.  A global network is built by combining all expression measurments (microarray samples) into a single compedium where all samples are normalized together and co-expression is determined through correlation of expression across all samples. 
  2. Gene Interaction Layer (GIL).  Unlike a global network where all samples are used to build a single network, a GIL is contructed by first clustering the input samples by expression similarity.  This clustering yields n groups of samples.  Each group is then passed through the network construction pipeline to form n unique networks. Global networks cannot represent relationships that occur in only a few samples (experimental conditions) and can be heavily biased towards a specific condition when it is over-represented.  Pre-clustering os samples is a mechanism to avoid those issues and capture as many meaningful interactions as possible.

 

What are the Different Types of Modules?

Modules are subsets of a network that contain highly connected nodes.  Typically genes within a module are involved in similar biological processes.  Functional enrichment analysis is one tool used to help identify potential function of modules.   There are  many different algorithms for determining modules within a network.  The modules in this site have been constructed using  WGCNA , MCL and link-community (LCM) methods. 

 

Can I find Gene Set's Causal for Specific Phenotypes With This Database?

This database does not provide a set of genes as targets for phenotypes, but it will allow you to explore and potentially find gene sets that may be involved in a specific phenotypes.   After exploration and identification of a potential gene set further tests in the laboratory would be needed to validate the gene set.

Copyright © 2010-2013. Feltus Lab, Clemson University