To date, taxonomists have identified about two million of the estimated ten million species of plants and animals thought to live on Earth.
Schindel noted that, using DNA bar codes, scientists could highlight species that have no match in a database of bar codes, freeing up taxonomists "to focus their attention on things that are new and exciting."
Bar Code Meeting
Next month scientists will meet in London, England, for the consortium's first international conference. There, they'll discuss progress in DNA-bar-coding science and kick-start international collaborations to create a bar code database for Earth's estimated ten million species.
The database will be housed at GenBank, the United States National Institutes of Health's publicly accessible database of DNA sequences.
In addition, Hebert and his colleagues have already begun to compile a bar code database at the University of Guelph. Discussions are being held to make sure the two databases are complementary, Schindel said.
To make the bar code database an economical, effective, and useful tool, the consortium is focused on a portion of a single gene common to all life, known as CO1, or cytochrome c oxidase 1. Variations in this segment of DNA distinguish species in much the same way that bar codes in a supermarket might distinguish brands of cereals.
"What we have said is that the CO1 gene seems to work very well for most animal groups," Schindel said. "So if you're going to use bar codes, start with that one."
The CO1 gene is found in mitochondria, the energy-producing subunits of cells that are inherited only from an organisms' mother. (Most DNA is found in the nuclei of cells and is inherited from both the mother and father of an organism.)
Hebert, the Canadian zoologist, said mitochondrial DNA accumulates mutations up to ten times faster than nuclear DNA. As a result, the mitochondrial DNA of closely related species differs more than their nuclear DNA, making it easy to tell the species apart.
"It's not just rapid rates of evolution but rapid divergence," Hebert said. "Nuclear DNA retains old variation, while old mitochondrial variants are much more rapidly lost, making species separation very straightforward."
The CO1 gene is also simple to isolate in a variety of animals, and a broad range of animal life has been shown to have distinct CO1 sequences, Hebert said. However CO1 does not work well for vascular plants (plants with tube systems for conducting fluid, such as ferns and flowering plants).
Scientists will discuss possible plant genes that can be used to quickly identify species at the February conference.
Proof of Concept?
Two studies published in October 2004 illustrate the power of DNA bar codes.
In a PLoS Biology paper, researchers led by Hebert sequenced DNA bar codes for 260 species of North American birds. Each had a unique bar code. Variation between species was, on average, 18 times higher than between individuals of the same species.
During the course of the study, Hebert and his colleagues also found two divergent groups of CO1 bar codes within each of four species. This indicated that in each case, a single species was really twothe DNA bar codes revealed four new species of North American birds.
Meanwhile researchers led by Daniel Janzen, a biologist at the University of Pennsylvania in Philadelphia, used DNA bar codes to show that a common butterfly in Costa Rica, Astraptes fulgerator, is actually ten different species. The study was published in the Proceedings of the National Academy of Sciences.
The researchers were tipped off to the possibility when they realized that more than 2,500 wild-caught A. fulgerator caterpillars could be separated into several groups according to slight variations in color and food preferences. The adult butterflies, however, are indistinguishable.
"Given the vast range of the supposed A. fulgerator species, it doesn't take much imagination to realize that there are probably a great many more hidden species out there," Janzen said in a media statement. The butterflies are found from the southern United States to northern Argentina.
Vertebrate zoologists Craig Moritz and Carla Cicero of the University of California, Berkeley, commented on the Hebert-led study of North American birds in PLoS Biology. There, the pair echoed the major criticism against DNA bar codes, questioning whether the technique can reliably distinguish between closely related species.
Mortiz and Cicero wrote that DNA bar coding, used in combination with other taxonomic practices, can help to identify individuals and to increase rates of species discovery. But the pair remained skeptical on when and where the technique will be useful, noting that "the real challenge lies with tropical taxa. "
Schindel, of the Consortium for the Barcode of Life, said his organization is open to such criticism of bar coding. But he does not believe bar code limitations should prevent taxonomists from having the technique in their toolboxes.
"I think the most important question is, Does DNA bar coding contribute to the process of taxonomic revision and refinement? And I think the answer is very clearly yes," he said.
SOURCES AND RELATED WEB SITES