User Tools

Site Tools


Comparing co-expression networks from different datasets

Category:Data Analysis
Supervisor:Raivo Kolde
Abstract:It has been shown several times that by using co-expression networks one can discover functional gene modules. However, often the modules that come out of the data are very similar, regardless of the dataset used. Therefore, it would be interesting to study how similar or different are the the networks that are based on different expression experiments.


  • Generate few networks using same parameters but different datasets -for example using correlation distance
  • Characterize and compare obtained networks by standard network properties, like node degree distribution, …
  • Compare networks in Graphweb, to see if there are functional groups either general in all or specific to some networks.
  • Compare also with some PPI networks


  • Reimand et al. GraphWeb: mining heterogeneous biological networks for gene modules with functional significance. Nucleic Acids Res. (2008) pp.
topics/bib/kolde2011b.txt · Last modified: 2011/10/19 10:34 by swen