Untangling the intricacies of cancer biology using computers can identify treatment targets that wouldn’t otherwise have been considered, according to research by an Australian team.
Professor Mark Ragan from UQ’s Institute for Molecular Bioscience led the team, which undertook an analysis of different computational methods of studying gene regulatory networks.
“Cancer is not a disease caused by single genes. Rather, it is changes to the underlying gene regulatory networks that prompt tumours to grow and spread,” Professor Ragan said.
“Understanding gene regulatory networks in healthy and diseased tissues is therefore critical to devising effective cancer treatments.”
These networks involve vast numbers of interactions between different molecules, making conventional experimental approaches, which are focused on individual genes, too time-consuming.
By contrast, computational methods can examine complex networks of interacting molecules across entire systems. The challenge for researchers is determining the accuracy of such methods.
The IMB team undertook a thorough analysis of nine different computational methods that represented a variety of approaches. They then applied the method judged most effective to real ovarian cancer data.
“Our evaluation demonstrated that it’s possible in some cases to use computational methods to gain insights into cancer biology.
“These methods can pinpoint targets that wouldn’t otherwise have been considered, which can then be validated with laboratory experiments.”
The findings are published in the current edition of the scientific journal Genome Medicine, where it has been nominated as part of the thematic series “Cancer bioinformatics: bioinformatic methods, network biomarkers and precision medicine”.
Subscribers can access the paper at:
http://genomemedicine.com/content/4/5/41/abstract and view the series at:
http://www.biomedcentral.com/series/cancer_bioinf
Media contacts
Professor Mark Ragan – 07 3346 2616 or m.ragan@imb.uq.edu.au
IMB Communications – 07 3346 2134 or 0418 575 247
