"Sequencing is expensive, so the shorter the section you can sequence and still get useful information, the better," van der Lelie said. "In fact, because these tags are so short, we 'glue' 10 to 30 of them together to sequence all at one time, making this a highly efficient, cost-effective technique."
For tag sequences that can't be matched to an already sequenced bacterial genome (of which there are only a couple hundred), the scientists can use the tag as a primer to sequence the entire attached ribosomal gene. This gene is about 1400 genetic-code-letters long, so this is a more time-consuming and expensive task. But since ribosomal genes have been sequenced and cataloged from more than 100,000 bacterial species, this "ribotyping" technique makes use of a vast database for comparison.
"If there's still no match," said van der Lelie, "then the tag probably identifies a brand new species, which is also very interesting!"
In another test with possible applications for identifying agents used in bioterror attacks, the technique also clearly discriminated between closely related strains of Bacillus cereus, a pathogenic soil microbe, and Bacillus anthracis, the bacterial cause of anthrax.
This technique could also help assess how microbial community composition responds to changes in the environment. Such information might help identify which combinations of species would be best suited to, say, sequestering carbon or cleaning up radiological contamination.