CONTOUR V1

NEW!! Inferring synthetic lethal interactions from mutual exclusivity of genetic events in cancer (2015):
Srihari S, Singla J, Wong L, Ragan MA (2015). Inferring synthetic lethal interactions from mutual exclusivity of genetic events in cancer. Biology Direct, to appear.



NEW!! An up-to-date survey of protein complex prediction methods (2015):
Srihari S, Yong CH, Patil A, Wong L (2015). Methods for protein complex prediction and their contributions towards understanding the organisation, function and dynamics of complexes, FEBS Letters doi:10.1016/j.febslet.2015.04.026. [PDF with slides!]



Systematic tracking of dysregulated modules identifies novel genes in cancer

Sriganesh Srihari and Mark A. Ragan
Institute for Molecular Bioscience
The University of Queensland
St. Lucia, Brisbane
QLD 4072, Australia

Email: { s.srihari | sriganesh.m.s } at { uq.edu.au | gmail.com }






The CONTOUR workflow (Jump to see diagrammatic illustration of each step)

Citation: Srihari S, Ragan MA (2013). Systematic tracking of dysregulated modules identifies novel genes in cancer, Bioinformatics 29(12):1553-61. [PDF]


Supplementary materials

1.
CONTOURv2 software release (written in C/C++) 
What can CONTOURv2 do ? (see figures below for illustration)
  • Generate conditional PPI networks: Given a PPI network and gene-expression samples from different conditions (e.g. normal, cancer sub-type, etc.), it can generate conditional PPI networks taking into account the original PPI weights (if any) and evidence from mRNA co-expression among protein (gene) pairs.
  • Identify complexes from PPI networks: Uses a maximal-clique merging method to identify complexes from the conditional PPI networks.
  • Match complexes between conditions: Given complexes extracted from two different conditional PPI networks (e.g. normal and tumor), it can match these complexes and identify those that show considerable changes in co-expression or composition in their constituent proteins. This can be useful to identify dysfunctional/disrupted complexes in cancer that potentially harbor cancer genes or targets.
Bug reporting: Please report all bugs in the code to Sriganesh. The typical errors could be due to size restrictions or memory allocations some of which are hard-coded for the sake of ease. These can be immediately fixed.


2. Cytoscape files:



3. Suppl. files (PDF) to the paper.



Citation:
Srihari S & Ragan MA (2013) Systematic tracking of dysregulated modules identifies novel genes in cancer, Bioinformatics 29(12):1553--1561.








Diagrammatic illustration of different steps in CONTOUR:

a. Module-extraction from PPI networks
  • Extract maximal cliques
  • Merge highly overlapping and interacting cliques to generate complexes





b. An illustration of how changes in relative ranking of maximal cliques affects module generation (between Normal and Tumor) and identifies gene-switching.






c. Matching of modules across conditions (Normal and Tumor)
(for CS enthusiasts, this is a maximal weighted matching in a bipartite graph)







 


Acknowledgments
We thank Professors Sean Grimmond (IMB) and Kum Kum Khanna (QIMR) for valuable discussions, and Dr Sarah Song (IMB) and Piyush Madhamshettiwar (IMB) for the gene expression datasets used in this study.
Thanks to Dr Guimei Liu (NUS) whose CMC clique-extraction and merging C++ routines were reused as part of this project.

Funding:
 Australian NHMRC grant to Dr Peter Simpson and MAR.





Check out our latest work BoolSpace!!
visitor statistics