Google goes cancer: Using the search engine’s algorithm to find cancer biomarkers
The strategy Google uses to decide which pages are relevant to a search is being used to determine which proteins in cancer are relevant for the progression of the disease. Researchers from Dresden University of Technology used a modified version of Google's PageRank algorithm to rank 20,000 proteins by their genetic relevance to the progression of pancreatic cancer. They found seven proteins help assess how aggressive a patient's tumor is. The Google strategy takes into account the content of pages and how they are connected by links. With this as a model, the authors made use of the fact that proteins in a cell are connected through a network of physical and regulatory interactions. "Once we added the network information in our analysis, our biomarkers became more reproducible," said Christof Winter, the paper's first author. This allowed them to find overlap with an earlier study and a connection was made with the proteins that can assess cancer aggressiveness.