Forgot your password?
typodupeerror
Google Medicine Technology

Researchers Use Google's Search Algorithms To Fight Cancer 52

Posted by samzenpus
from the searching-for-a-cure dept.
MatthewVD writes "German scientists have modified Google's PageRank algorithm to scan tumors and learn more about how cancers progress. PageRank orders results based on how other web pages are connected to them via hyperlinks; the modified algorithm, NetRank, scans how genes and proteins in a cell are similarly connected through a network of interactions with their neighbors. This approach could also yield new therapies to help combat tumors."
This discussion has been archived. No new comments can be posted.

Researchers Use Google's Search Algorithms To Fight Cancer

Comments Filter:
  • by Anonymous Coward

    This is just another markov model, just like PageRank. Move along people, nothing to see here.

    • Exactly. The funny thing is, markov model is also one of the basic things used to game Google's algorithm itself, with the autogeneration of content.

      Which makes me wonder, on the all things that interest geeks, why do they ignore things like communication? Why is there no comprehensive research done towards generation human readable text by automation?
      • by Anonymous Coward

        Which makes me wonder, on the all things that interest geeks, why do they ignore things like communication? Why is there no comprehensive research done towards generation human readable text by automation?

        Yourself make seem generation almost readable text by automation. Will you confirmation Turing test on the soon?

      • by Anonymous Coward

        Why is there no comprehensive research done towards generation human readable text by automation?

        There is. It's just very difficult to generate useful text, moreso automatically. Markov chains give you good results (considering the little effort required) but the text is gibberish. I don't think there's a "killer application" for this, though. Find a way to kill people with it and sure enough we'll have it in a few years.

      • by Anonymous Coward

        Economics mostly. It's still cheaper to hire a bunch of people from China/India/Alabama or other third-world location. Generating text by automation seems to require a pretty strong AI, and that's hard to get.

    • That... is the weirdest definition of Markov model I have ever heard, but I guess Wikipedia agrees with you.
  • The technique these researchers and Google Search use is known from the 1950s. Google didn't invent it. Just as Steve Jobs didn't invent the smartphone, Mark Zuckerberg didn't invent social networks, and Bill Gates didn't invent the PC.
    • Doesn't matter. They have the patent. That's all that counts.

      So, expect any future cancer cures to either be blocked from the market, or served with craploads of ads.

  • by Rogerborg (306625) on Friday May 18, 2012 @06:56AM (#40039407) Homepage
    • Researchers: "[NetRank operates] in a manner similar to Google's PageRank"
    • Retard masquerading as a professional journalist: "Researchers modified PageRank to develop NetRank"

    Die. Just die in a fire. Die, die, die.

    • by durathor (450646)

      Seriously, did you RTFA where the researcher himself described the algorithm as being based on PageRank

      “We first experimented with our own ideas on network algorithms until we realized that what we needed existed already with the PageRank algorithm, so why reinvent the wheel?” Winter recalled.
      “Our PageRank-based algorithm singles out proteins in the cancer cells that seem to either promote or suppress disease progression,” Winter said.

      How about the abstract of the research paper in q

      • by Kurrel (1213064)
        PageRank is designed around the 'random surfer' model (15% chance to jump to a random webpage) and returns the probability that a surfer will arrive back at the same page after a large number of links. A node that only has two edges can easily have the highest PageRank.

        If all they want is to find highly-connected nodes, then the Oracle of Bacon [oracleofbacon.org]-style average length solution would work just as well.
    • Researchers: "[NetRank operates] in a manner similar to Google's PageRank"
      Retard masquerading as a professional journalist: "Researchers modified PageRank to develop NetRank"

      Die. Just die in a fire. Die, die, die.

      Researchers' paper: "NetRank is based on Google's PageRank algorithm."

      Read. Just read the article. Read, read, read.

  • by Anonymous Coward on Friday May 18, 2012 @07:15AM (#40039487)

    The idea is interesting but the paper seems flawed. They integrate data and then look at how genes are positioned in the network with page-rank. The problem is that genes positioning in the network is highly dependent on how studied they are. Therefore, very well studied will get a high "NetRank." Genes known to be predictive of cancer progression are very well studied (lots of fudning in that area). This means the algorithm is basically finding and returning a list of what we already know, and it turns out that what we know is reasonably predictive when you combine 400+ markers.

    I'm surprised this made it by peer review without additional experiments to assess the role that this bias plays.

  • The thigh bone's connected to the leg bone; the leg bone's connected to the ankle bone...
  • look at the name, it says it all .. its a ranking algorithm.

  • Look Google we get it you still take your whole 'don't be evil' thing serious but I think you might be brown nosing a bit now.
  • by Errol backfiring (1280012) on Friday May 18, 2012 @11:16AM (#40041401) Journal
    Colour me surprised if cancer cells use Google to fight back.
  • That should make this guy [slashdot.org] happy.

% APL is a natural extension of assembler language programming; ...and is best for educational purposes. -- A. Perlis

Working...