When clinical researchers in a Boston hospital were looking for a better way to diagnose inflammatory bowel disease (IBD) in children, Neil Rasmussen ’76 got involved. He helped fund a project to look for diagnostic clues in the microbiome, the complex ecosystem of microbes that lives inside the human body and supports human health. MIT biological engineer Eric Alm, the Karl Van Tassel Career Development Associate Professor, was already involved, developing a statistical model that collects stool samples from patients, analyzes the genomes of the microbes present, and recognizes differences from healthy samples.
Rasmussen immediately recognized the value of connecting clinicians to engineers. “MIT was an obvious nexus for this, but instead most research was happening in isolated pockets,” he says. “I saw huge value in trying to glue that together, to create a critical mass and get more creative people collaborating in this emerging field.”
Rasmussen took action through a $25 million gift to fund the Center for Microbiome Informatics and Therapeutics, a strategic partnership between MIT and Massachusetts General Hospital (MGH). The Center will sponsor collaborative science between clinicians trying to help their patients, and engineers and scientists who collect and make sense of the volumes of patient data needed to study their conditions.
Ultimately, the Center is committed to understanding the role of the microbiome in human diseases such as obesity, autism, rheumatoid arthritis, multiple sclerosis, and other autoimmune disorders, with an initial focus on IBD. “Many of these diseases seem to have correlations in the microbiome,” says Alm, co-director of the Center along with Ramnik Xavier, MD, PhD, chief of gastroenterology at MGH. “If we find correlations, then we may be able to develop diagnostics, and in some cases those correlations might give us ideas on how to develop therapeutics to treat disease.”
Finding those correlations is a Big Data problem, according to Rasmussen. Really big. “Every person’s microbiome is unique. It would be one thing if the microbiome was comprised of three kinds of bacteria, but there are thousands,” he says. “And the deeper we dig, the more we find.”
The first step towards solving this problem is to collect data, lots of it, from lots of patients. In studying cancer, researchers have learned that new drugs that target specific biological functions often work extremely well for a very small subset of patients. Finding those patients requires large numbers of participants in clinical trials and the analysis of reams of data to determine which patients benefit. This approach is also expected to be useful for diseases rooted in the microbiome.
Microbiome research adds to that complexity because it involves not only analysis of human genomes, but also the genomes of the myriad microbes that support human biology, and the interactions those microbes are having with the human immune system and metabolism. Sequencing and informatics tools have advanced to the point where this work is feasible, says Rasmussen, who is a non-voting member of the Center’s steering committee. What is needed now is high-quality data.
As a first step, the Center is engaging with clinical researchers who are studying IBD to help them develop standards defining what data to collect and how best to collect it. Such standards, the first application of Rasmussen’s “glue,” will allow researchers to make the most of the valuable samples collected from patients so that, together, the entire ecosystem of microbiome researchers can learn as much as possible from it.