Researchers Invent “Google” for Disease

A research team lead by scientists at the Broad Institute has developed a new tool that uses unique genetic patterns to link specific illnesses with potential treatments.

Called the Connectivity Map, the development has already yielded insights into potential treatments for cancer and Alzheimer’s disease, the team reported in the Sept. 28 immediate early edition of Cancer Cell and the Sept. 29 issue of Science magazine.

The researchers have likened the approach to an Internet search engine like Google, which is able to sift through masses of data to find the best match between the attributes of a drug and the symptoms of a disease.

“The analogy is pretty good. In our search box, you can upload a list of genes, and with one click, the analysis is computed,” said Justin Lamb, the lead author of the Science paper and a senior scientist in the Broad Institute’s Cancer program. “The philosophy is also very much the same as a search engine like Google. Our program is free, and it is a tool that can allow scientists to generate information cheaply and fast.”

To build the Connectivity Map, the scientists described the effects of different drugs and diseases using the language of “genomic signatures”—the set of genes that respond to a particular drug or disease with expression level changes. The scientists compiled the genomic signatures of more than 160 drugs and other biologically active compounds.

One key finding suggests a new way to overcome drug resistance in cancer. A scientific team led by Scott A. Armstrong, assistant professor of pediatrics at the Medical School and an investigator at the Dana-Farber Cancer Institute, identified the Food and Drug Administration-approved immunosuppressant drug, rapamycin, as a therapeutic candidate for overcoming drug resistance in a form of human leukemia.

Lamb believes that the map can accelerate the search for new drugs to treat diseases with the support of the scientific community.

“Based on how successful the map seems to be with only 160 drugs, increasing the number of drugs for which we have the necessary information would increase the power of the map greatly,” Lamb said. “We already have the infrastructure in place to expand this resource quickly.”

The Connectivity Map marks another step in an ongoing boom in biomedical genetics research. Earlier this month, a team including Will Parsons, a researcher at Johns Hopkins University and the National Cancer Institute, sequenced the genetic code for human breast and colon cancers.

“Often medicines target particular genetic receptors, rather than giving us any idea about the background complexity of cancer,” Parsons said. “Computational advances in studying gene expression have really allowed us to scale things up and understand the pathways behind groups of genes with similar functions.“

—Staff writer Anupriya Singhal can be reached at asinghal@fas.harvard.edu.