Question of the Fortnight Can Microsoft really cure cancer?

A new Uk-based team is treating the disease like a computing problem

It’s a big claim for a company that took 13 years to get rid of a talking paperclip, but Microsoft is planning to cure cancer within 10. The tech giant has set up a ‘biological computation group’ in Cambridge, where programmers and engineers will work alongside oncologists and biologists to tackle the disease.

The group’s most ambitious project is to build ‘molecular’ computers that can be placed inside cells and programmed to detect and destroy tumours. Andrew Phillips, who leads the group, doesn’t underestimate the challenge, calling it a “moonshot” on Microsoft’s News website ( www.snipca.com/21973, see image). But he says it should be “technically possible” within five to ten years.

Dr Jasmin Fisher, senior researcher and associate professor at Cambridge University, was similarly bullish, saying: “I think for some of the cancers five years, but definitely within a decade. Then we will probably have a century free of cancer”. She talks not of curing cancer, but “solving” it, like you would any other computing problem.

Those involved see the research as a logical extension of computing.

Chris Bishop, laboratory director at Microsoft, told Wired magazine: “I think it’s a very natural thing for Microsoft to be looking at because we have tremendous expertise in computer science and what is going on in cancer is a computational problem”.

He said that while biology and computing may seem like very different disciplines, they “have very deep connections on the most fundamental level”. Bishop added: “The complex processes that happen in cells have some similarity to those that happen in a standard desktop computer”.

Microsoft’s first step is to write software. The Cambridge group has already developed code that mimics a healthy cell. They weren’t the first team to do so, though – that honour goes to Stanford University in the US, where a complete software model of a bacterium was created in 2012.

The next step is to come up with a code that lets the team find the differences between healthy and cancerous cells. Making the tiny computers to be implanted into cells is the final part of the process, but it could also be the most difficult. If Microsoft succeeds, it will be the biggest breakthrough in the field of molecular computing since the idea was first proposed 40 years ago.

‘Reprogramming’ tumours is not the whole project, however. The other half of Microsoft’s plan is to apply techniques such as machine learning to the massive amount of data that exists about cancer. This may be less headline-grabbing than the innovative research under way in Cambridge, but it could achieve important results.

The group is already working on a system that could help doctors interpret a 3D scan of a tumour. By analysing pixels in the scan they’ll be able to work out how much a tumour has

shrunk, grown or changed.

All this research could mean cures for at least some cancers in the next 10 years. But Microsoft may not be the first to find them. The field of computational biology has grown rapidly since Stanford’s pioneering work four years ago, with other labs, such as the one at the New York Genome Centre, delivering promising results. Such vibrant competition can only be a good thing in the battle against cancer.

In 10 years tiny computers may destroy tumours in our bodies