Skip Content

Google’s code helps to identify pancreatic cancer biomarkers

18 May 2012

Researchers at Dresden University of Technology, Germany, have modified the way that the search engine Google calculates and ranks the popularity of web pages to identify biomarkers for pancreatic cancer.

The research, published in PLoS Computational Biology, ranked 20,000 proteins by their connection to the progression of pancreatic cancer and found seven that can help to assess how aggressive a patient's tumour is.

Biomarkers are molecules produced by cancer cells that can help to detect cancer earlier and guide decisions on treatment, such as whether or not a patient should receive chemotherapy. Identifying biomarkers is difficult and often studies on the same cancers don’t find the same biomarkers.

This problem has been avoided using the Google method, which takes into account the content of a web page and also how these pages are connected via hyperlinks. With this strategy as the model, the authors made use of the fact that the many proteins in a cell are connected through a network of complex interactions; a 'protein Facebook', so to speak.

"Once we added the network information in our analysis, our biomarkers became more reproducible," said Dr Christof Winter, the study’s lead researcher.

The team validated the biomarkers on 412 samples taken from pancreatic cancer patients who had surgery between 1996 and 2007. The team’s biomarkers were more accurate than established clinical factors predicting outcome, such as grade of cancer and tumour size.

When validated further, these biomarkers could be used as a way of diagnosing pancreatic cancer. In addition, they may also help in the development of new drugs to slow down the progression of cancer. 

In the hope of developing new treatments, the team is collaborating with the Dresden-based biotech company, RESprotect, which is running a clinical trial on another pancreatic cancer drug.

< Back