New 'C-Map' Links Drugs to Diseases In a paper published in Science magazine, a group of scientists introduce the Connectivity Map (C-Map). It's an online database connecting the genetic signatures of diseases to those of available drugs. The map could significantly speed up the rate of drug discovery, and find new uses for old drugs.

New 'C-Map' Links Drugs to Diseases

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript



A bit later, we'll be talking about drug safety. But first, you know the Human Genome Project and the sequencing of all the genes in our genome, and researchers hoped that this information will lead to a better understanding of diseases and treatments.

When the sequencing was finished, scientists cautioned that, you know, the really hard work lay ahead - and that was mining all that information, turning the information into useful products. Well, something new called a Connectivity Map is a tool that delivers on that premise. In a paper published in today's Science magazine, a group of scientists are introducing the C-Map. It's an online database connecting the genetic signatures of diseases to those of available drugs. And by combining the data from the human genome with the information from the diseases, the Connectivity Map could significantly speed up the rate of drug discovery and perhaps find new uses for some of the old drugs that are already out in the market.

Joining us now to talk about the Connectivity Map is one of its creators. Todd Golub is director of the cancer program at the Broad Institute of MIT and Harvard and an investigator at the Dana Farber Cancer Institute in Boston. Welcome to the program, Dr. Golub.

Dr. TODD GOLUB (Broad Institute of MIT and Harvard; Dana Farber Cancer Institute): Good to be with you.

FLATOW: Thank you. Why is it so hard to match diseases with effective drugs?

Dr. GOLUB: Well, the usual way to approach this is to study each disease in great molecular detail until you understand it well enough to match it up with a particular drug. But in most cases, that's very difficult to do and we don't even know where to start. So the concept here was to try to use the human genome to try to systematically put the language of diseases and the language of chemicals, or drugs, into a common language and that is a language of the human genome.

FLATOW: Tell us more about this common language or their common ground and why that's useful?

Dr. GOLUB: Well, as you know all cells whether they're deceased or normal, or whether they're treated with a drug or not treated with a drug, contain the same genes. They have the same genome sequence. But when we treat cells with a drug or when a cell becomes deceased, certain genes turn on or off. And we can think of this as a gene expression signature.

So it occurred to us, and in particular, Justin Lamb - one the scientists in our group at the Broad Institute - that one could define a gene expression signature of diseases and one could define a gene expression signature of cells treated with drugs, thereby having the same language to connect drugs with diseases, and thereby identify connections between diseases and drugs.

FLATOW: You can measure many things in a cell to get a signature. Why use genetics?

Dr. GOLUB: Well, it's a very good question. In principle, one could measure all sorts of different things. But there are advances in technology that allow one to measure the activity of every gene in the genome all at once, very rapidly, and at only modest expense. And so this seem to us to be the ideal method using the so-called DNA chips to measure lots of things about a cell, very rapidly.

FLATOW: Talking to Todd Golub about the Human Genome Project and using new methods of possibly using old drugs. Let's get into that in a minute. 1-800-989-8255 is our number. So tell us in - we now know the basis for this new genetic roadmap. Tell us about that. How big is it? Is it really - is it a map that dots get connected by lines and things like that.

Dr. GOLUB: Well, it's perhaps not a map in the usual sense of thinking of a map, but it's a map to the extent that one can use this information to figure out, for example, given a drug - what other drugs are nearest to it? And one can learn about the activities and the actions of a new drug for example, by seeing what else is near it. And so in that sense it's a map.

Or one could think about a particular disease and say what other diseases or molecular processes, or drugs for that matter, are nearby. And so it's the map analogy, I think, is useful there as well.

FLATOW:` So you create a database of all the different reactions of the drugs and the diseases and you farm that database? You search through it looking for connections?

Dr. GOLUB: That's right. A concept that was really defining of the Human Genome Sequencing Project and pertains to the Connectivity Map also, is a notion that all of this information about our genome should be available for free to the public, for scientists to use in useful ways. So that's exactly right. The notion of the Connectivity Map is that there would be a large database of genetic profiles of drugs that scientists and researchers throughout the world could mine over the Internet to develop new ideas for connecting drugs to disease.

FLATOW: Let's talk about the nuts and bolts of how this might work. Let's say I have the genetic footprint of a disease and I want to match it to a drug. What do I do?

Dr. GOLUB: Well, you would do a - what is now a very standard experiment in a laboratory, or research laboratory - to define a signature or fingerprint of the disease that you care about. And then you would take that signature, which is simply a list of genes in the genome that are turned on or off in the disease that you care about, and you would submit that through a Web site, over the Internet, to the Connectivity Map database. And the computer would churn in the background and compare the signature that you put in, to all the signatures that have been put into the database, and return an answer , do… Does anything in the database match what you put in? And if so, what was it?

FLATOW: You've likened this process to a Google search of genetic footprints. Explain that analogy.

Dr. GOLUB: Well, I think the analogy to Google is again the notion of having something freely available to anyone who has a computer, and an approach that enables a user to define what is searched. So in the Google case, there might be something that you're looking for on the Web. In the case of the Connectivity Map, the user puts in a signature of a disease of interest, and then look through all of the data bits in the database to find something that best matches. And does that in a very transparent way, that's without restriction of use.

FLATOW: So this is an open source sort of thing?

Dr. GOLUB: It is to the extent that the database, we believe, should be freely available to anyone.

FLATOW: Can people add to the database?

Dr. GOLUB: Well, at the moment, we're conceiving of this in a way where the data are centrally generated so that we can systematically keep control over the way in which the data are generated. But it's conceivable that in the future, that indeed others may join in this effort. And indeed, the next phase of the project will likely include others joining in.

FLATOW: Well, how about going in the opposite direction. Will the Connectivity Map help a scientist find or discover potential new uses for drugs we know about already?

Dr. GOLUB: Well, that's one of the most exciting potential applications of the Connectivity Map. As you know it takes a very long time for a drug to pass through all of the safety testing that's required before it's reasonable to make the drug available to patients. And what scientists are discovering is that drugs that we've been giving for many, many years to patients for various diseases, often have more than one effect in a cell, but we haven't had a way to discover those before. So the hope is that tools like the Connectivity Map will allow scientists to discover new, previously unrecognized functions of all drugs that we know are very safe but might now use for new types of conditions.

FLATOW: Equally so, could you not now take a new prescription drug that's in development and run it through the database and find perhaps unwanted side effects from it?

Mr. GOLUB: Well, that's a possibility. It's one that we haven't explored in as much detail. Finding unwanted side effects is a complicated undertaking in that some of those side effects can be very common in many individuals and happen right away. Those are probably going to be easier to detect in this type of approach. Those that are occurring later in individuals and are very rare in small percentages of patients are probably going to be more difficult to detect in this way.

FLATOW: Does the genome project genes that were used - Will the genome give you a wide enough assay of the general population to discover all these things?

Mr. GOLUB: Well, there is minor variation in the genome sequence between individuals, but most of what a cell does is the same in a normal cell, in a diseased cell or in cells from different individuals. So the ability to measure, for example, all of the genes in the genome at once I think is giving us a very rich snapshot of what's going on in a cell in a way that will be generalizable.

FLATOW: What else can you do with it? I'm sure you thought of other interesting things for a database like this.

Mr. GOLUB: Well, part of a challenge of the Human Genome Project is, having identified what all the genes are in the human genome, we still don't know what most of them do. And so one hope is that we could use the Connectivity Map to connect the action of genes in the genome to diseases they might be involved in by finding a signature of a disease and a signature of what a cell looks like when we experimentally turn on or off a particular gene. If we see a match in the signature, that would suggest that maybe this gene that no one was ever thinking about before, has something to do with that particular disease but no one thought to consider that possibility.

FLATOW: So when does this go online? When will it be in the public domain?

Mr. GOLUB: Well, the public database is online now and it's available for scientists to use now, and we expect that this database will grow dramatically over the next year or two.

FLATOW: Well, we'll all be watching and waiting for the details to see what happens. So what's the next phase? Can you expand or what happens? Where do you go from here for what's next for the C Map?

Mr. GOLUB: Well, I think the most important next step for us is to develop gene expression signatures for all drugs that have been already approved by the FDA. These are most exciting because if scientists could discover new uses for old drugs, one could imagine moving those very rapidly into the clinical trial testing phase. So we're most motivated to expand the database in the direction of having profiles of every drug ever approved by the FDA.

FLATOW: Wow. Can you do that? I mean, it would take some time I imagine - a lot of drugs.

Mr. GOLUB: Well, our goal is to accomplish that over the next year to two.

FLATOW: Just the year to two? Wow.

Mr. GOLUB: The technologies have advanced sufficiently so that one can do this sort of work robotically much more rapidly than one could have even considered a year ago.

FLATOW: I'll bet. Thank you very much for taking time to talk with us.

Mr. GOLUB: My pleasure.

FLATOW: Todd Golub is director of the cancer program at the Broad Institute of MIT and Harvard and investigator at the Dana Farber Cancer Institute in Boston. We're going to take a short break and continue our drug theme this afternoon. We're going to talk about prescription drug safety, a new report that looks at whether the FDA is up to the task of making sure all these new drugs that come out on the market are safe once they get on the market, so stay with us. We'll be right back after the short break.

(Soundbite of music)


Copyright © 2006 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.