Microsoft wants to "solve" cancer.
The company has a secretive 150-person medical arm in Cambridge, U.K., known as its "biological computation" unit. According to Fast Company, that team has the goal of creating living cells that can be programmed to treat diseases the way computers can be programmed. Chief among those diseases: cancer.
Achieving this end first requires the ability to be able to map the many interactions that take place within a cell. The Microsoft team specializes in biological modeling, which involves using advanced, software-based computation to achieve this. It's an emerging field that first gained wide recognition in 2012, when a team at Stanford successfully created a computer model of a complete parasite.
The next part of the effort is controlling the way cells behave. "We can use methods that we've developed for programming computers to program biology, and then unlock even more applications and even better treatments," said Andrew Phillips, the head of the Microsoft biological computation unit, in a publication put out by the company.
That requires developing a programming language that can mimic the interactions that occur naturally within a cell. If that can be done, then scientists could create or restrict certain cell behaviors--like the ones that cause cancer. Should the earliest stages of cancer be detected, the cell could be programmed to fight and halt the disease.
Another team within the Microsoft unit, led by neuroimmunologist and Cambridge associate professor Jasmin Fisher, studies how sick cells behave compared to healthy ones. Fisher's group is developing a cloud-based platform that can allow scientists to easily study the millions of interactions taking place within a person's body--including the ones that cause cancer--and develop individualized treatment for that person. In this way, Fisher believes her team could solve or "debug" cancer.
It all sounds far fetched, but experts seem to think these outcomes are possible--someday. The relationships that take place within a cell are extremely complex and are the result of millennia of evolution. Creating computer systems that can fully understand and recreate them will take time.
For now, Fisher's team has partnered with pharma giant AstraZeneca on a project to help scientists easily understand which patients are likely to develop resistance to specific drugs. The company would then be able to suggest alternative drugs to overcome the resistance, as well as determine which therapies are worthwhile for a particular patient.