Using PageRank to analyze the spread of cancer cells
As we have been discussing in class, Google’s PageRank algorithm is used to rank webpages by Google, but a few researchers from the University of Southern California, Scripps Clinic, the Scripps Research Institute, the University of California, San Diego Moores Cancer Center and Memorial Sloan-Kettering have applied the same algorithm to analyze patterns in the spread of cancer cells in the body. These researchers found that cancer cells move around the body in more than one direction simultaneously. Additionally, they also found that certain organs spread cancer cells more aggressively than the others and that some organs act as sponges of cancer cells.
PageRank can be understood as a Markov chain where the states are the various webpages and the transitions are the equally probable links between pages. These researchers tried to translate this model to the spread of cancer cells where the various states (previously webpages) are the various organs or parts of the body and the transitions are the possible paths for the spread of cancer cells from a specific organ to another. The weights of these paths are obtained from the combined autopsy data of 163 cancer cases.
The above diagram is a representation of the network diagram that is formed where every possible organ/ body part is a node and every edge is a possible path that the cancer cells could use to spread through the body.
Therefore, the researchers used the concept of PageRank, which was created for the purpose of finding more relevant information on the World Wide Web, and applied it to the problem of tracking and analyzing cancer cells in the body. Another interesting point that the author of the below article stated is that the advancements in computing and data storage technology have improved the ability to handle large datasets such as the one used by the researchers in this case. In the future, the same “quantitative and computational framework” could be applied on “larger, better and more targeted datasets” to get more useful results about specific genotypes or phenotypes.[1][2] Lastly, the use of algorithms, such as PageRank, that have been developed with computer science applications in mind, in fields other than computer science fascinates me. The possible widespread applications of the concepts taught in this course have excited me a great deal as well.
[1] https://gigaom.com/2013/03/26/how-researchers-are-fighting-lung-cancer-using-pagerank/
[2] http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0034637