Continuous medical research has led to the availability of a myriad of information relating to almost limitless topics, but the challenge has been organizing all of the data in a way that facilitates its usability. Prof. Shai Shen-Orr and researchers from his lab at Technion – Israel Institute of Technology’s Rappaport Faculty of Medicine have spent years creating an artificial intelligence computer software called the immuneXpresso. The software scans through millions of published scientific journal articles and interprets the data that is collected into maps of disease profiles. The technology also uses machine learning to suggest previously unknown biological interactions that might be relevant for a specific search. In one particular study, the immuneXpresso was programmed to scan the PubMed database for the search term “cytokines proteins,” whose function is sending messages between immune cells throughout the body. The program created a map of the body illustrating the work of the cytokines based on the existing understanding of the proteins and also suggested new theories about them. The technology can be practically applied not only to give educated hypotheses regarding linkage of various diseases to the immune system, but also, when it is supplied with a given individual’s immune profile, can create a personalized medicine or treatment plan.
The research has been supported by the National Institute of Health and was published in the journal Nature Biotechnology. Tel Aviv-based Cytoreason, the largest systems immunology group in the world, has taken over the development of the program in an effort to create a commercially available machine learning model of the immune system that will help pharmaceutical companies deliver new drugs and precision medicine.