Eran Segal
Eran Segal is a professor in the Department of Computer Science and Applied Mathematics at the Weizmann Institute of Science in Israel, heading a lab with a multi-disciplinary team of computational biologists and experimental scientists in computational and systems biology. His group has extensive experience in machine learning, computational biology, and analysis of heterogeneous high-throughput genomic data. His research focuses on the microbiome, nutrition and genetics, and their effect on health and disease. His aim is to develop personalized medicine based on big data from human cohorts.
When did you become interested in microbiome research?
I am a keen marathon runner, and as such I developed a strong interest in nutrition and how the microbiome responds to fitness diets. I originally started out in a different area of computational biology called gene regulation, but as my own interest in nutrition grew, I realised that the microbiome was a rapidly evolving area of new research in which I was keen to get involved. We had been working on a personalized nutrition project using our own novel immunology approach and this provided us with a ready-made collection of samples from our cohort, and the technological basis for this new study.
What is your project about?
We already know the importance of both the immune system and the microbiome to health. We know that the immune system and the microbiome interact; every day the human body produces around two grams of antibodies against gut microbes. Keeping the body in a healthy state requires a fine balance between the microbiome and immune responses. However, we have been largely blind to the scope of these immune-microbiome interactions because of our inability to perform measurements and analyse them at scale. Typically, a research project will study one particular gene from the immune system and examine its interaction with one specific bacterium. Our project will, for the first time, provide comprehensive mapping of immune-microbiome interactions in both healthy individuals and patients with autoimmune and auto-inflammatory conditions. We hope to determine which immune responses to gut-based antigens might trigger or exacerbate disease. We have created a pioneering high-throughput approach that can methodically identify hundreds of thousands of molecular interactions using a combination of machine learning, bacteriophage display technology, and robotic automation. It should provide detailed profiles of the way the immune system and microbiome interact under healthy and diseased states.
Segal’s project will comprehensively map the interactions between the microbiota and the immune system.
How does your novel screening approach work?
From 1,000 microbiome samples, taken from 250 patients with inflammatory bowel disease (IBD), 250 with multiple sclerosis (MS), and 500 healthy controls, we will select bacteria that are particularly abundant or interesting. We then select specific antigens within those bacteria; molecular regions that are likely to be recognized by the immune system, such as secreted proteins and membrane proteins. Hundreds of thousands of these antigen designs are entered into a computer, and sent away to be synthesized. We use these synthesized antigens to create a bacteriophage library, wherein each phage displays a specific antigen and holds the DNA from which the antigen was produced. Next, we take blood samples from people, each of which contains that person’s unique set of antibodies, and incubate each blood sample with the phage library. Any antibody that recognizes a specific antigen will bind to it. We can then extract all the bound antigen-antibody complexes, and subject the phage DNA to next-generation sequencing. This shows us which DNA elements are enriched compared to the initial library we started with. Enrichment tells us to what degree that particular antigen was selected by the antibodies of that person. We have worked very hard to calibrate our system carefully and increase the signal to noise ratio. The technology is very robust. It can test up to one million antigens against a given blood sample in one experiment, and it can run up to 100 samples simultaneously.
What insights will your results yield?
For each individual, the process will reveal which microbial antigens their immune system recognizes, and then we can then correlate that with other health data we hold on that individual. This will allow us to pinpoint direct relationships between IBD, MS and microbial activity, and hopefully highlight potential biomarkers for these conditions.
We are trying to answer broad questions. For example, is there a relationship between your current microbiome composition and what your immune system recognizes from these bacteria? Perhaps your immune system is actively excluding certain bacteria; it recognizes the bacteria as a danger or a source of intolerance, or it is taking note of which bacteria are safe and are part of your commensal microbiome.
To give a specific example, leaky gut is a process implicated in conditions such as obesity and diabetes. When a person suffers from a leaky gut, bacteria escape from the intestine and reach other parts of the body. With luck, they are stopped by the immune system. When that process does not fully stop, the infiltrated bacteria cause inflammation. If a person has, or has had, a leaky gut, we believe there will be a signature present in the immune system that our technology can find. It should also indicate the magnitude of the signal, providing information about the seriousness of the individual’s condition.
What will be the long-term applications of your findings?
Firstly, our work will contribute to current scientific knowledge and advance our basic understanding of immune-microbiome interactions. Secondly, our study should identify biomarkers that could help develop valuable assays for diagnostics, prognostics and for measuring the efficacy of interventions in different diseases. The immune system signatures we find could be relevant for developing novel therapies. Ultimately, this platform could help with personalization of medicine, by highlighting antigens that individuals cannot tolerate with a simple, efficient test. The technology we have developed is very broad, it can test your immune response against any antigen.
