Using information from patient records at different stages of inflammatory bowel disease, New York scientists have developed the world’s first gene-based predictive model of the disorder.
An article on the model, titled “A functional genomics predictive network model identifies regulators of inflammatory bowel disease,” was published in the journal Nature Genetics.
Previous studies had identified more than 200 genes that play a role in IBD, but the data was not in a form that would allow a research team to create a model that could predict the disorder’s development and progression.
Researchers at the Icahn School of Medicine at Mount Sinai in New York and other academic institutions used information on three groups of patients at different stages of their disease to design the computer-based predictive model.
The model takes into account the complex network of biological pathways involved in IBD’s immune response. This means its key focus is mechanisms that regulate the disease.
In addition to patients’ treatment records, researchers used information on their DNA variations, gene expression, and the mechanisms that regulate their disease to create the model. Expression is the process by which a gene creates a functional product such as a protein.
“These results demonstrate how much we stand to gain by organizing massive amounts of molecular and clinical data using advanced machine learning approaches that in turn can be queried to generate novel disease insights,” Eric Schadt, the Icahn School of Medicine’s dean for Precision Medicine, said in a press release. He was the lead author of the study.
“Our predictive model serves as a repository of knowledge and understanding that facilitates learning more about the development and progression of IBD, including identifying master regulators of disease that can be explored as targets for treatment,”said Schadt, who is also CEO of Sema4, a partnership of scientists, doctors, engineers, and genetic counselors whose aim is to use data to improve health.
Analyzing the information on patients’ genes, the mechanisms that regulate their diseases, and how the two interact also allowed researchers to create an IBD immune-regulatory network. The team used experiments to validate the top 12 genes in the model, a process that generated new insights into the disease’s regulatory mechanisms.
“By creating multiscale predictive models of the immune component of IBD across different stages of disease, this work helps move us towards a more comprehensive understanding of the complex molecular network of this disease,” said Scott Snapper, a Harvard Medical School professor who is also chair of the Crohn’s and Colitis Foundation’s National Scientific Advisory Board. “I look forward to accessing this new knowledge base,” he said, adding that it will be “highly informative to the collective efforts of the IBD research community to identity new targets and ultimately novel treatments of IBD.”