Systems Biologist and Data Scientist Shai Shen-Orr is the Co-founder and Chief Scientist of CytoReason and a Professor in the Faculty of Medicine at the Technion—where he directs the laboratory of Systems Immunology and Precision Medicine. Shai is devoted to developing new analytical methodologies for grappling with the intricate complexities of the immune system and their application in the search for new answers to disease and drug development. Shai’s research laid the foundation for formation of CytoReason, a company that has developed a unique immune system-focused artificial intelligence approach to generate biological insights to help drive drug development. (Source: TedMed)
The following has been paraphrased from an interview with Prof. Shai Shen-Orr on July 12th, 2018.
(Click above to listen to the full audio version or click here for a downloadable version)
Could you walk me through how the field of systems immunology took off?
The immune system is extremely complex and traditionally it has been studied either by isolating different cells, one at a time, to understand what each does, or by examining whole organisms, usually a mouse, and inducing some disease to see its effects. That started to change around 2007 when new technologies allowed us to measure the breadth of the immune system from a single blood draw by measuring multiple cell types simultaneously. The biggest breakthroughs came in 2010 when we suddenly had technologies that allowed us to measure a vast amount of cell types all at once and see the relationships between them. This made the challenge of taking drugs from mice to humans much smoother. But the field is still young, while our ability to measure the immune system has greatly improved, we are lagging in our ability to make sense of all those measurements to understand how all the components work together.
Did the field experience any push-back from the traditional reductionist approach to biology?
Yes. This is a revolution that dramatically changed how people work and it has not been an easy transition for many. It requires a different type of education, particularly more emphasis on data science so one can be comfortable with the computational analysis and machine learning techniques used. There was push-back, but that is diminishing as we have more success and as the need develops to study certain immunotherapies and other new drugs on a systems level.
How complex is our immune system compared to other complex systems like the Earth’s atmosphere or macroeconomics?
It is difficult to quantify. Every cell is different and there are hundreds of different cell types with different functions. Also, in every individual they change over time. Even if you have twins, born with the same DNA, their immune systems are going to adapt and change as they are inevitably exposed to different things in our environment. It is an extremely complex system; I would probably rank it second behind only the brain in human physiology.
How complete is our understanding of the immune system?
I would say that we know we are still relatively far off. We recently built a natural language processing engine (artificial intelligence technique) that can read all of the papers on immunology and quantify what the different cell types are and how they relate to one another. What we realized is that about 17% of the communication cells use accounts for 50% of the current knowledge in the field. Combined with all we know from human and mouse trials, we estimate that at best we understand about 1/8th of the system and even this depends on the resolution you choose to look in and what you would call ‘understand’. However, the good news is that even with just that small amount of information we are having success in identifying biomarkers and drug responses.
Is our immune system centrally controlled or does it act as a hive mind?
For the most part it is a distributed system that is not controlled by any one center. But, we are now looking more at the relationship between the brain and the immune system. What we are seeing is that there is information that the brain passes on to the immune system that does influence it.
Some talk about immunology as the answer to everything from cancer to neurodegenerative disease, what do you think are the upper limits of what immunology is capable of?
I believe its role is still underestimated, it is the system in charge of sensing our environment and protecting us from it. That is both internal, as in cancer, and external, as in infectious diseases. We also keep finding new diseases in which it is involved, most recently it was found to affect cardiovascular diseases, as well as Alzheimer’s and Parkinson’s disease. Also, it is networked with all of our other biological systems, so it can have very broad impacts.
How far do you think we are from understanding why our immune systems degrade as we age and will we be able to prevent this from happening?
As we age our immune system does lose some of its ability to protect us. It could come from an adaptation in the system that has a cost, but that still remains to be seen. In my lab, and in collaboration with Mark Davis at Stanford, we have been tracking the immune systems of humans for 9 years to see how they change as people age. I would say these changes are due to adaptations. Because it is an integrated network, changing any one feature does not resolve some of the issues that the system faces as it ages and that makes it difficult to effectively harness the immune system to tackle chronic diseases. We still have a lot to learn before we’ll get there.
20 years from now, what will immunology enable us to do that we can’t do today?
Our understanding of the immune system is growing all the time, and the company I co-founded, CytoReason, is already building on this knowledge with its technologies in order to make important breakthroughs in disease mechanisms, target identification, biomarker discovery and indication selection. Looking further ahead though, what we are starting to get a glimpse of is precision immunology – using a combination of drugs, each of which modulates the system, matched to the specific patient and their specific disease. We should also see some approaches that allow us to delay the aging of the system. Today precision medicine just uses our understanding of genetic pathways to develop targeted treatments. I think in the future targeting the immune system will be much more beneficial as it is easier to examine and has a greater impact on our health than our genetics.
Prof. Shai Shen-Orr’s lab at the Technion
Handy infographic on CytoReason
Another interview with Prof. Shai Shen-Orr
Press Release on Immune-Focused AI Model
Press Release on machine-learning-facilitated biomarker discovery