Some worms eat in company. Fear, rather than fellow feeling, may drive them to dine together, research on new-found nervous system genes suggests. That a few genes can have such a profound effect on behaviour provided food for thought this week at the First Annual Symposium on Neurogenomics, held in Chapel Hill, North Carolina.

Let loose on a plate of Escherichia coli bacteria, British roundworms (Caenorhabditis elegans) scatter and munch alone. German and Australian strains bunch and eat together. A tiny change in one gene, encoding a cell receptor, transforms worms from loners to social animals, Cornelia Bargmann and her colleagues at the University of California in San Francisco told the conference's 500 delegates.

Using gene chips holding dots of 2,100 nervous system genes, the group compared other genes that were active in friendly and loner worms -- and spotted 73 differences. "There seems to be a global change in the nervous system," says Bargmann.

The genes affect confidence: to a worm, bacteria smell frightening. "Worms eat E. coli only because they’re forced to -- they smell terrible," explains Bargmann. When thrown on a plate of reeking bacteria, the ‘sociable’ worms bunch together because they feel threatened; loner worms are more confident.

Of the 73 gene differences, many work in synapses, the connections between nerve cells, Bargmann and her co-workers have found. Together, they may alter how the worm reacts to smells of both foetid food (bacteria) and pheromones (smell signals from other worms).

"There are complex traits in worms as in people," says Bargmann. She believes that her work illustrates how the study of genomes can shed light on human behaviour. "You can take something as initially fuzzy as a behavioural pattern and analyse its [molecular] components," says Bargmann.

Before, neuroscientists studying a behaviour or disease would study one gene for years: now, with high-throughput genomics methods, they can quickly screen thousands. A bottleneck has been removed, says Stephen Strittmatter, who works on nerve regeneration at Yale University School of Medicine. "It opens up new techniques."

Using genomics means "the pace of change has really picked up", Gerald Fischbach, a neurologist at Columbia University, agrees. "The brain is the most complex object in the known Universe -- we’re not going to know its deepest secrets in this millennium."