The same tools you use for looking up cat videos might also offer scientists new insights into cancer, Jason Bittel writes this week in Slate.
Highlighting a recent paper in the journal Cancer Research, Bittel describes how researchers at the University of Southern California are using an algorithm "similar to Google PageRank" to better understand patterns of lung cancer metastases.
Using data from old lung cancer autopsy reports, the researchers are using Markov chains to work backwards from this data to develop models of how metastatic lung cancer spreads.
As USC researcher Paul Newton tells Bittel: "Basically we're doing the inverse of what Google does. They know the transition probabilities and compute the steady-state, we know the steady-state and compute the transition probabilities."
Using this Markov chain-based analysis, the team has identified regions of the body – such as the liver, lymph nodes, and bones – that appear to trap and halt the spread of metastases, and regions – such as the kidneys and adrenal glands – that appear to promote their proliferation. Bittel notes that these findings could offer "a new way to target treatments and lessen or prevent the spread of cancer."