You will not die alone, but with a trillion-strong audience of bacteria and other microscopic buddies. They are glued to our tongues, swarming in our intestines and hitching a ride in our noses. They aid our digestion and shape our immune systems.

Delving into these orifices and their microbial residents is exciting many microbiologists — a sentiment apparent in certain sessions of the American Society for Microbiology meeting in Orlando late last month. Keen to document exactly which microbes are there, many researchers are using brute-force ‘metagenomic’ studies in which they extract microbial DNA from the body cavity of choice and hurl it all into a sequencing machine. They can identify hundreds or thousands of microbial genes in one swoop.

The discussions in adjacent rooms offered a sharp contrast. Clinical microbiologists — those tasked with identifying the meaner, infectious microbes that invade our bodies and hospitals — use a different set of time-honoured techniques.

Conventionally, they take a swab or sample and attempt to identify the culprit microbe by growing it on certain agar media or examining the shape of a fungi's tentacles under the microscope. Genetic tests are not routine and are typically done to identify one organism at a time. Results can take a day or two to come through. Doctors facing a severe infection use pre-emptive antimicrobial drugs that might later prove fruitless.

Clinical microbiologists have good reasons for sticking with these seemingly old-fashioned techniques. Above all, they are foolproof, cheap and definitive — why use an expensive sequencing machine when a culture plate will do? Genetic tests run a high risk of artefacts because one stray contaminating bacterium could produce a false result — and clinical microbiologists have only one precious sample and cannot tolerate errors when a patient's life is at stake.

But used alongside culture methods, the potential rewards of sophisticated genetic analyses are tantalizing. They can reveal a host of bacteria that we cannot grow in culture. Some researchers are designing gene chips that could, if adapted for use in hospitals, reveal which of a range of microbes is there, what strain it is, and which antibiotic-resistance genes and virulence factors it harbours. Used widely, such profiling could help show where a microbe has come from and whether it is spreading.

There is gap to be bridged here. Basic researchers have created ever more crafty ways to hunt down microbes. Clinical microbiologists, who have to deal with hundreds of different diagnostic tests simultaneously, want robust, high-throughput methods that have clear benefits in helping them choose a drug or curtail an infection. As is so often the case, research that converts the former into the latter is in short supply — and neither group has the time or expertise to do it.

More cash for this type of research would, as always, help. It could be used to train more clinical microbiologists in advanced molecular techniques or equip laboratories to trial new diagnostic methods. Meanwhile, basic and clinical researchers can make a simple start, by talking to each other a little more, rather than sitting in separate conference rooms.

Genetic snapshots of the teeming mass of bodily microbes complicate things further. Take, for example, a study released last month into the bane of cystic fibrosis sufferers, Pseudomonas aeruginosa (E. E. Smith et al. Proc. Natl Acad. Sci. USA 103, 8487–8492; 2006). By tracking one infection over eight years, the study showed that this bacterium is a moving target: at any one time, many different lineages are present, presumably nestled into different crevices of the lungs. In order to understand and halt an infection, it may not be enough to know that P. aeruginosa is there — we might need to know all the different variants and how they interact with each other and with other microbes. This area is ripe for research and might reveal a chink in the microbe's antibiotic-resistant armour.

Metagenomic studies are fascinating for those of us grown-ups still childishly interested in our own dank, dark recesses. But microbial surveys have urgent application in medical care. The research and medical community must ensure they are used there.