Microsoft wants to “solve” cancer, and is doing it by thinking about the body like a computer.
The technology giant may be more closely associated with malware than malignant diseases, but researchers working for the company’s “biological computation” unit in Cambridge are showing the former isn’t entirely separate from the latter.
“The field of biology and the field of computation might seem like chalk and cheese,” Chris Bishop, laboratory director of Microsoft Research’s Cambridge lab, told Fast Company. “But the complex processes that happen in cells have some similarity to those that happen in a standard desktop computer.”
Bishop and his team of scientists are working on ways to treat living cells like programmable computers. Overlapping the fields of machine learning, biology and computer-aided design, the project aims to build molecular computers, built from DNA, that work within human cells and automatically look for anomalies such as cancer, and then eliminates them.
Andrew Philips, who leads Microsoft’s bio-computing group, told The Telegraph: “It’s long term, but… I think it will be technically possible in five to ten years’ time to put in a smart molecular system that can detect disease.”
While the long-term goal for Microsoft’s biocomputational group is to create biocomputing systems capable of detecting and treating diseases, in the short term researchers are modeling processes that happen within cells – to analyse where problems start and how they can be stopped. This information is already being used by pharmacologists, who want to gain a better understanding of why certain patients form resistances to drugs.
“I’m hoping this is the beginning of changing the way we do drug discovery,” AstraZeneca’s Jonathan Dry told Fast Company. “We could… test all our ideas in a system like this, and determine the experiments that will have the best chance of success.”
You can read a full account of Microsoft’s biocomputing efforts in the company’s extensive blog post on the subject.
Disclaimer: Some pages on this site may include an affiliate link. This does not effect our editorial in any way.
