Plant species could soon be identified from tiny crumbs of leaves or seeds using databases of standard DNA "barcodes" — a tool that until now was only available for animals.

If the DNA segments endorsed by an international group of 52 scientists Monday do become standard, they could lead to handy tools for non-scientists such as border agents and people who sell traditional medicines.

There is a huge diversity of plants and many look so similar that even experts may have trouble identifying them, said Sujeevan Ratnasingham, a University of Guelph researcher who worked on the study.

"And of course, it can be downright impossible if you've only got a small piece of plant material," he said. "Standardization changes everything — it democratizes the process. It means that anyone who has that standard can go out and get an identification."

A standard portion of DNA is already used as a barcode to identify 60,000 animals and counting, but scientists had been previously unable to agree on a suitable standard barcode for plants.

In a study published Monday in the Proceedings of the National Academy of Sciences, scientists from more than 20 institutions in the U.K., Canada, the U.S., Mexico, South Korea and about half a dozen other countries endorsed a set of two-plant DNA segments to use as standard barcodes.

Ratnasingham said barcode identification could allow customs officials to easily detect invasive plants at the border, and could also be used to conduct research to set different types of government policies. For example, researchers could analyze the gut of an insect to see whether it is eating a commercially valuable crop on a large scale.

Or, suggested University of Guelph biology professor Mehrdad Hajibabaei, it could be used by natural health practitioners or government inspectors to identify a mixture of plant material marketed as a traditional herbal medicine.

7 sequences, 400 plants

The researchers came up with the standard sequences after examining seven that had been proposed in recent years, including some that had been suggested by Hajibabaei and co-authors in earlier studies. They analyzed those sequences from more than 400 land plant samples.

Ratnasingham, whose job involves the computational analysis of DNA data to extract the information coded inside, said they were trying to choose a standard segment based on two criteria:

• It needed to give a clear, unambiguous signal using standard methods.
• The DNA sequence had to be consistently similar among members of the same species, but consistently different among different species.

The study found that using two segments of DNA found in chloroplasts — the part of a plant cell responsible for photosynthesis — researchers could identify the plant 72 per cent of the time. In the rest of the cases, they could tell that the plant was one of only a few closely related species.

The standard means only a small amount of DNA information needs to be collected from each plant in order to compare it with thousands of others, said Hajibabaei, who tested and sequenced samples sent to him from the Smithsonian Institution in the U.S. and the Royal Botanic Gardens in the U.K. Having a minimalist barcode is much faster and easier than trying to sequence the entire genome of each species.

Ratnasingham said the study tried to be as inclusive as possible so that people at research institutions, regulatory bodies and at companies around the world will adopt the standard and start incorporating the segments into barcode databases. He added that those segments have already been entered into the Barcode of Life database he is involved with at the Biodiversity Institute of Ontario.

Other researchers from the University of Toronto, the University of British Columbia and the University of Guelph were also involved in the PNAS study.