Researchers from Dresden University of Technology, Germany have adapted the algorithm used by Google's web search engine to develop a computational programme that can identify panels of biomarkers (molecules produced by cancer cells) for cancer prognosis.
The study was published in PLoS Computational Biology on 17 May 2012.
Google's PageRank algorithm takes into account the network of hyperlinks between web documents as well as the search terms themselves to determine which pages are most relevant to a particular search query. The researchers adapted this in order to develop NetRank, which uses biological interaction information between genes' products to better decide which genes are the most relevant for outcome prediction. They applied NetRank to gene expression profiles obtained from 30 patients with pancreatic cancer, and identified seven candidate marker genes.
The hope is that the method can quickly help identify the proteins that assess how aggressive a tumour can be, and help doctors decide whether to provide chemotherapy or not.
Until now, finding these biomarkers has been difficult and time consuming - and the markers vary for all the different varieties of cancer, and even vary within the same forms of cancer.
The researchers have been able to rank around 20,000 proteins by their genetic relevance to the progression of pancreatic cancer and are now working with Dresden-based biotech company RESprotect, who are running a clinical trial on a pancreas cancer drug.
The full study can be found on the PLoS Computational Biology website