Researchers have completed the first-ever genome-wide scan of synesthesia, a condition in which sensory stimuli cross wires and combine such that people "see" sounds or "taste" shapes, according to a study published online today in the American Journal of Human Genetics.

Investigators at the Wellcome Trust Centre for Human Genetics at the University of Oxford pinpointed four areas of the genome associated with the disorder. Those regions contain genes that have been associated with autism and dyslexia, as well as genes involved in different aspects of brain development, and further analysis could illuminate how genetics drives complex cognitive traits, the authors say.

"It's exciting that we have a study about the genetic basis of synesthesia -- finally," said Noam Sagiv, a cognitive neuroscientist at Brunel University in the UK, who was not involved in the research. Until now "we've just been guessing," he said, by "using data based on prevalence estimates."

Francis Galton, best known as one of the founders of eugenics, first pointed out in the 19th century that synesthesia runs in families, suggesting it has a genetic component. "Based on pedigree analysis, many people thought that [synesthesia] was a simple Mendelian disorder," said Julian Asher, now at Imperial College, London, the study's lead author and himself a synesthete. The study, however, demonstrated that the condition is multigenic. Which makes sense, Asher noted: "The clinical phenotype is so complex, it's hard to believe all of that can be caused by a single gene."

Consequently, "now we're looking at gene-gene interactions, with multiple genes contributing," Asher said.

The condition comes in several forms, but Asher and his colleagues conducted their genetic analysis on individuals with what's known as known as auditory-visual synesthesia -- the classical form, in which individuals strongly associate sounds with specific colors.

The region of the genome most strongly linked to synesthesia was on chromosome 2, and has also been strongly linked to autism. That doesn't mean that the two conditions are related, per se, explained Ed Hubbard, a cognitive neuroscientist at Vanderbilt University in Tennessee who was not involved in the study. Instead, the common gene or genes are likely "more generally involved in how the brain gets built."

The study also pulled out a region on chromosome 6 that contains genes linked to dyslexia -- especially interesting, "seeing as phonemes [the units of sound in language] and letters are two of the strongest synesthetic triggers," Asher said. One possibility, he said, is that mutations in those genes lead people to process those stimuli differently. "Maybe some people [with those mutations] end up with dyslexia, and some with synesthesia."

The genetic analysis also dispelled the long-held assumption that synesthesia was an X-linked disorder, an idea that stemmed from the fact that researchers had never before seen synesthetic fathers who had synesthetic sons. In their subject pool, the researchers identified two cases of such male-to-male transmission, demonstrating that the condition is almost certainly autosomal.

Asher and his colleagues are following up their study with higher-density screening, to narrow down their list of candidate genes. "Next, we want to home in on what these genes are and what they do," he said.

At least two other labs, one at Baylor College of Medicine in Houston, and the other Trinity College, Dublin, have been working on similar genetic studies, but the subject database of those groups differs from that used in the current study, said Hubbard. The Irish group includes fewer families, but with more individuals from each family, and the Houston group focuses on people with a form of the condition called sequence-to-color synesthesia.

The differences in the populations "could make a huge difference," said Sagiv. Asher and his colleagues were wise to choose a very specific population of synesthetes for the field's first genetic study, he said, but broadening the inclusion criteria to different forms of the disorder may identify additional genes.

Hubbard agreed that the results of all three genetic studies together will further flesh out the picture. "Once we start to identify genes that are involved, we can start to look at what function they play in a mouse or monkey brain," he said, and then "come up with ways of studying what they do experimentally" in animal models.