Google Algorithm Links to Cancer Detection
PageRank, the algorithm used by Google to measure the relative importance of websites on the Internet, could be used to shed light on the organizational principals of other complex network systems, such as the human body, say Northeastern University researchers.
World-renowned network scientist Albert-László Barabási, and Gourab Ghoshal, a postdoctoral research associate in Barabási’s lab at Northeastern, recently analyzed the revolutionary link analysis algorithm used by the popular search engine.
The findings were published in the July 19 issue of the journal Nature Communications.
PageRank, researchers found, can be far more than the back-end tool used by Google to help web geeks find the latest Internet meme in record time. In cellular networks, PageRank could help scientists identify proteins that play important roles in cancer.
As Ghoshal put it, “You can examine how proteins interact with each other and then identify genes based on their PageRank value that are important in determining cancer.”
The effectiveness of the PageRank algorithm depends on the stability of the structure within which data are ranked.
“Had the World Wide Web not followed the structure that it does, then PageRank would have been much less effective in providing us with the most pertinent web pages,” said Barabási, a Distinguished Professor of Physics with joint appointments in biology and the College of Computer and Information Science, and the founding director of Northeastern’s Center for Complex Network Research.
The latest paper builds upon earlier research featured in the May 12 issue of Nature, in which Barabási and his postdoctoral research associate, Yang-Yu Liu, examined the ways in which greater control of complex systems such as cellular networks or social media, can be achieved by merging the tools of network science and control theory.
In June, Barabási was honored with the Institute for Scientific Interchange Foundation’s 2011 Lagrange-CRT Foundation Prize for his body of research on complex networks in natural, technological and social systems.