The Google of Drug Discovery?

New tool aims to link diseases with treatments.

In September, researchers in Boston launched what they hope will become the Google of drug discovery, a free, genetic search engine they call the “Connectivity Map.” One goal is to match diseases with potential treatments, another is to suggest how drugs can be applied in new ways to treat diseases like cancer.

To build a searchable bridge between drugs and disease, researchers had to find a common language, says Justin Lamb, a lead researcher for the project and a senior scientist at the Broad Institute of MIT and Harvard. “Our decision then was to represent disease and drugs using the language of gene expression,” Mr. Lamb says.

To create the Connectivity Map, the team measured the genomic signatures of 164 drugs, then designed a computer program to compare the signatures of the drugs with each other, and with the signatures seen in diseases.

Using signatures as one would use keywords with Google, users can explore how a compound might impact a disease. And so far the map has yielded some interesting finds.

Mr. Lamb cites, for example, the discovery of a drug that can help therapy-resistant patients fight a fast-moving form of leukemia called acute lymphoblastick leukemia. The drug was sirolimus, an immunosuppressant, Mr. Lamb says. Made by Wyeth Pharmaceuticals, sirolimus is normally used to help prevent the body from rejecting organ transplants.

Researchers at the Dana-Farber and Children’s Hospital in Boston are now planning to trial the drug.

But their map is still some ways off from acquiring Google-like status. The database is hardly encyclopedic at this point. Although researchers plan to pump it up with the signatures of every FDA-approved drug, about 1,400, Mr. Lamb says more work will be needed to determine how useful the Connectivity Map will be.

The map is not the first searchable database in this area. Mountain View, California-based Iconix Biosciences, for instance, uses similar technology to help big pharma garner a better understating of a potential compound’s toxicity. And as odd as it may sound, Iconix salutes the arrival of the Connectivity Map. “I think it’s a great validation for the approach,” says Don Halbert, Iconix’ vice president of R&D.

That said, there are big differences between the two search engines. The Iconix engine harnesses the ability to search and understand compound toxicity in rats. The Connectivity Map looks at gene expression using human cells. As such, Mr. Halbert says the map doesn’t pinpoint how an entire organism’s system might react to a drug.



Related Links:

• Go back to category Internet & Networks to read more reviews and articles.
• Go back to Software Magazine's main page.