杏吧原创

听

Gay or straight? A saliva test can predict the answer, and get it right 67 per cent of the time 鈥� for male identical twins at least.

The test, which uses clues from tiny modifications to a person鈥檚 genome, is the first that claims to detect sexual orientation. Many scientists have expressed caution over the results, while concerns over potential misuse of the test have led the study鈥檚 lead researcher to quit the project entirely.

鈥淭he scientific benefit to understanding [why people vary in sexual orientation] is obvious to anyone with an iota of curiosity,鈥� says at Northwestern University in Evanston, Illinois. 鈥淭he predictive test needs replication on larger samples in order to know how good it is, but in theory it鈥檚 quite interesting.鈥�

Over the last two decades, several studies have suggested that sexual orientation is, in part, down to our genes. Perhaps the biggest splash was made in 1993 by Dean Hamer鈥檚 team at the National Cancer Institute in Bethesda, Maryland, when they found that gay brothers tended to share a sequence of five genetic markers in a region of the X chromosome. The same region has been implicated in other studies of sexual orientation since, although researchers haven鈥檛 been able to single out 鈥済ay genes鈥�.

Other observations also suggest a genetic basis for sexual orientation, such as the mysterious fraternal birth order effect. For every male pregnancy a woman has, a subsequent son has a 33 per cent higher chance of being homosexual, although no one knows why. The overall chance is still low, however, rising from around 2 per cent to just 6 per cent for a third son.

鈥淚t seems as though the mother鈥檚 body is remembering the sex of previous pregnancies,鈥� says at the University of California Los Angeles. A male pregnancy might leave some sort of marker behind that affects subsequent pregnancies. This might be down to epigenetic changes 鈥� the addition or subtraction of a methyl group to genes, which switches them on or off.

To investigate a potential role for epigenetics, Ngun and his colleagues looked for epigenetic modifications made to the genes of 47 sets of male twins. Ten听of the twin pairs were both gay, while 37 pairs differed, with one brother identifying as gay and the other as straight. They ended up with a giant spreadsheet showing the levels of methylation across the genome of each twin, says Ngun.

Next, Ngun and his colleagues looked at the genomes of homosexual and heterosexual volunteers. They used an algorithm to search out gene regions in which methylation patterns differed significantly between the two groups.

They found five sites鈥� three in regions of 鈥渏unk DNA鈥�, the role of which is unclear, two in genes whose roles are relatively well established.

One of these genes is involved with the production of MHC II molecules. These are important for a healthy immune system, but are also thought to affect sexual attraction.

The other gene region implicated is responsible for making a protein that affects neuron function. 鈥淚t could affect how neuronal circuits are patterned, and influence behaviour,鈥� says Ngun, who presents the findings today at the American Society of Human Genetics annual meeting in Baltimore, Maryland.

The team then went a step further. Using the test results from 20 of the pairs, they developed a model to predict if a person is straight or gay based on the methylation patterns of their genes. When they tested their model on the remaining pairs of male twins, they found it correctly predicted sexual orientation 67 per cent of the time.

False positives?

Other scientists are cautious about the results. 鈥淪tudies that associate biomarkers with particular traits are notoriously prone to false positive results due to the tendency of these studies to find spurious associations that are down to sheer chance,鈥� says Johnjoe McFadden, a molecular geneticist at the University of Surrey, UK.

Since the associations have not yet been tested in a completely independent study population, the results should be considered no more than suggestive. There needs to be verification before any firm conclusions can be drawn, he says.

While epigenetic patterns are showing promise as a biomarker for certain traits, it鈥檚 still difficult to infer causality. , who studies epigenetics at the Babraham Institute in Cambridge, UK, says: 鈥淭he nub of the problem with studies like these is that when you see methylation changes, you don鈥檛 know whether methylation is the prime event or if it鈥檚 reflecting some other event. Methylation might be reflecting a state rather than driving it.鈥�

Ngun himself has concerns that the test has the potential to be used and abused. 鈥淚鈥檓 gay,鈥� he says, 鈥渁nd I鈥檝e always wondered why I am the way I am. But once you have this information, you can鈥檛 control how it鈥檚 used or disseminated.鈥�

The study raises concerns that people could try to tinker with epigenetic modifications to change sexuality. Currently, we don鈥檛 have a way to selectively change epigenetic patterns on DNA, although the technology is being developed.

鈥淭echnically it might be possible to change DNA methylation in a targeted way in the near future,鈥� says at University College London.

This concern may be premature. Marc Breedlove at Michigan State University in East Lansing points out that in its current form, the test is not accurate enough to be used to predict whether someone in a new population of individuals is gay with any certainty, since the 67 per cent accuracy of the test is only relevant for the test population, who are themselves not reflective of the general population, in which a much lower proportion of people are gay.

Nevertheless, some researchers contacted by New 杏吧原创 raised concerns over the ethical implications of such research. For example, if the test were developed further, could it one day be used to screen for sexuality at an early age?

鈥淓ugenics is always a possibility, but governments that regressive would rarely have enough money to spend on something like this,鈥� says an ethicist and historian of sexuality. 鈥淢ore likely it would be used by parents.鈥�

Dreger recalls an anecdote from a researcher who studies the fraternal birth order effect. The researcher received a phone call from a man in the US who was looking to hire a surrogate mother 鈥� but because of the effect did not want someone who had already had several sons. 鈥淭hat鈥檚 not really what I want鈥︹�� the man had said, 鈥渆specially if I鈥檓 paying for it.鈥�

Should epigenetic changes be shown to be partly responsible for sexual orientation, it would still be difficult to screen for them. First one would have to work out at what age the marks appeared and whether they were permanent.

鈥淎ssuming the marks were placed early enough in fetal development, the potential for a [screening] test is there,鈥� says Ngun. But even so, it would be difficult to know which cells of an embryo to test, because epigenetic patterns vary between different types of cells in a single person. 鈥淲hich embryo cells would correspond to adult saliva?鈥� says Beck.

Ngun is concerned that his work could be misinterpreted by people who seek to identify and punish people for being gay. 鈥淚 honestly don鈥檛 think it鈥檚 that far-fetched,鈥� Ngun says. In some countries, homosexuality is punishable, sometimes by death.

The idea has troubled Ngun to the extent that he decided to abandon research in the field completely. 鈥淚 just left the lab last week,鈥� he says. 鈥淚 don鈥檛 believe in the censoring of knowledge, but given the potential for misuse of the information, it just didn鈥檛 sit well with me.鈥�

(Image: Masaaki Toyoura/Getty)