Photo Credit: Adaptive Biotechnologies

It would seem Microsoft are stepping up their genomics game this week. After we came to you yesterday with their suggested CRISPR improvement, today we come to you with yet more plans that effectively uses Artificial Intelligence (A.I.).

Microsoft Suggests Artificial Intelligence is the Answer to Improving CRISPR

Our bodies immune system inhabits a huge data problem. Through a new A.I. project, Microsoft hopes to tackle that by making diagnosing nearly any disease as simple as a single blood test, reports Gizmodo.

In a research effort with the Seattle biotech firm Adaptive, the company is working to decode the human immune system in order to diagnose disease. 

“Your immune system should know what you have before your doctor does,” explained Adaptive CEO, Chad Robins at the annual JP Morgan Healthcare Conference in San Francisco. 

The idea is to make a map of the human body’s immune responses, including its T-cell receptors sequences and the codes of the antigens they have fought against. Then, using that map they hope to diagnose practically any disease from a sample of blood. 

“We’re searching for patterns in a giant space,” commented Peter Lee, vice president of A.I. Research for Microsoft. “In machine-learning, a problem this big is exotic.”

Our bodies are always coming into contact with foreign invaders and having immune reactions to them. Due to the T-cell receptors binding to different antigens, the presence of one T-cell receptor could indicate a host of different diseases. As you can imagine, that’s a lot of data to sift through. What is promising however, is that all the information is there, we just need to figure out a way to read it. 

In order to begin sifting through the data, Microsoft and Adaptive need to sequence a whole lot of T-cell receptors, as well as to train the machine-learning algorithms. Although a universal diagnostic tool for disease is a long way off, a shorter term solution is within reach. 

“Sequencing your entire adaptive immune system is a huge data problem,” Robins added. “It’s like a giant jigsaw puzzle. It’s mostly a matter of doing the grunt work.”

Their first efforts will focus on hard-to-diagnose autoimmune disorders and infectious diseases, as well as high-risk cancers. Lyme disease has been suggested as a potentially good first candidate. The first diagnostic tools could even be as soon as three years away. 

“If we layer on diagnostic after diagnostic, ultimately we can get this to be a screen for the entire system,” Robins said. “We believe in not having this be some mega-moonshot.”

Lee suggested that the problem is very similar to that of machine translation. “We’re not able to read the antigens directly, but we can read the T-cell receptors to get this weird translation that your body has made of the disease-state,”he explained. “At a fundamental, algorithmic level, that’s very similar to problems we’ve already solved.”

Unfortunately, a fundamental problem when working on the human body is that there are always a number of unknowns. But, a universal map of the human immune system would allow doctors to diagnose patients earlier on, and with much greater accuracy – without putting them through an endless arrays of tests. Even more, access to such detail of a person’s immune system could also help predict how they might respond to new pathogens and treatments in the future.