CURRENT ISSUE		June 15, 2004

I.T. Indispensable to Human Genome Researchers

By Joseph Goedert, News Editor

In late 2000, gene researchers announced an enormous accomplishment the mapping of the human genome. The achievement came at least four years ahead of schedule because the federal government was pumping money into the project, researchers were excited and competitive, and advances in information technology improved computing power, says Steven Enkemann, manager of the micro array core facility at the H. Lee Moffitt Cancer Center and Research Institute in Tampa, Fla.

Our work today in science is absolutely impossible without I.T., he adds. Funding for the National Center for Biotechnology Information is the best money Congress ever spent on medical research.

The center, part of the National Institutes of Health, is a national resource for molecular biology information. It creates and maintains public databases storing genome data, develops software for analyzing the data, and serves as an information clearinghouse.

The center also developed the Basic Local Alignment Search Tool, or BLAST, an important advancement in database query capabilities, says Marshall Peterson, chief technology officer at the J. Craig Venter Science Foundation in Rockville, Md.

BLAST is a killer app in bioinformatics because it lets you look for similar characteristics in DNA, he explains. The same gene mutation that causes blindness in a fruit fly, for example, also causes blindness in humans.

Identifying mutations

It is the identification of these mutations of the tiny differences in a person's DNA or protein molecules that can pinpoint the cause of certain diseases, or whether a person is genetically predisposed to get a certain disease, or whether a person's body will respond to a specific drug therapy. A specific drug, for instance, significantly reduces tumors in lungs but only works for 10% of the population.

The human genome contains more than three billion bits of information, about 99.9% of which is the same in everyone. As researchers have learned more about what is the same and what is different, advancements in database technology have helped them collect and store huge volumes of data. It's those single changes in a gene that makes the number-crunching go up so much, says Kristin Newby, M.D., an associate professor of medicine at Duke University Medical Center in Durham, N.C.

Recent advances in database technology and other I.T. applications are letting scientists compare the genetic data of tens of thousands of people, isolate mutations, then identify commonalities in mutations. In recent years, software vendors have developed analytic databases designed to handle the complex query needs of gene researchers, Peterson says.

For instance, when one researcher is conducting a complex query on a transactional database, other researchers typically cannot use the database during that query period. This reduces productivity. The Venter Foundation, however, recently started using an analytical data warehouse from Framingham, Mass.-based Netezza Corp. that is specifically designed for biological research.

It replicates data from multiple transactional databases, Peterson says. This allows you to do queries on transactional databases while others are working on them.

The analytic data warehouse, called the Netezza Performance Server, also supports BLAST queries. It allows multi-function searches, such as look for all patients who are diabetic, more than 10% over their optimum weight, with silver hair, blue eyes and feet size 12 or larger, he adds.

Other technological advancements are speeding the process of identifying which characteristics of our genes bind us together and which set us apart. Photolithography technology, the same process that puts circuits on computer chips, now is being used to place different bits of DNA on tiny chips of glass. Light passed over the glass can help identify mutations. Previously, the limit was 10,000 bits on a chip of glass, but Santa Clara, Calif.-based Affymetrix recently introduced a two-chip set with 60,000 bits of DNA on each chip.

The Moffitt Cancer Center and Rochester, Minn.-based Mayo Foundation are migrating to the new chip technology, called the GeneChip Mapping 100K Array Set. It will make it easier to zero in on important genes, Enkemann says. It will be 10 times faster. In addition to research application advances, the maturing of the Internet and e-mail has had a huge impact on the ability of researchers across the world to collaborate.

The size of files we need to transfer are getting bigger and bigger, Enkemann says. You could fill the hard drive of a computer of 15 years ago with just one file that we transmit today.

The wish list

Gene researchers have much of the information technology they need to conduct their discovery-oriented work. What they lack is decision support software to make the work meaningful for clinicians, contends Kevin Schulman, M.D., director of the Center for Clinical and Genetic Economics at Duke University.

How do we bring the research data back to clinical care? he asks. I'm going to do all this elegant science but am still unclear on how to treat an individual patient. I can't remember three billion bits of genetic information on that person.

Researchers, for example, hope to identify gene profiles that indicate patients at risk for renarrowing of arteries following an angioplasty procedure. These patients are the best candidates for expensive drug-eluting stents that secrete a medication that prevents tissue growth from narrowing the arterial walls at the stent site. Further gene research will help identify which types of patients have a genetic makeup most receptive to the medication.

Bar code technology is helping researchers manage large numbers of test tubes holding individual samples of DNA and other genetic material. But the technology has limitations because of the difficulty of getting codes on small tubes. As research projects get bigger, such as a study of 500,000 people with 10 tubes needed for each patient, the scope of the management of these tubes becomes apparent.

Researchers are working with test tube manufacturers to find solutions, such as smaller bar codes or migrating to radio frequency identification technology, says Newby of Duke. It's not yet clear, though, if an RFID tag on a test tube can survive at minus 70 degrees for many years.