THE idea of telling if someone is gay or straight from their DNA is highly provocative. So no surprise that a study entitled “A novel predictive model of sexual orientation using epigenetic markers” has caused a big stir.

Focusing on genetic testing may attract attention, but as a scientist long interested in the role of the genome in sexuality, I am more intrigued by what the work tells us about the role of epigenetic imprinting – the silencing of genes by methylation. This imprint can pass from parent to child and has implications for a range of complex human traits.

The first DNA markers for sexual orientation date back to 1993, when my lab at the US National Institutes of Health found an area on the X chromosome – Xq28 – that influences sexuality in gay brothers. Earlier this year, that finding was robustly confirmed in a study 10 times the size.

Despite such work, it was clear that inherited differences in DNA could not account for all of the observed variation in sexual orientation. That led me and graduate student Sven Bocklandt to hypothesise in 2003 that epigenetics may play a role. We illustrated this by suggesting how atypical imprinting of the X chromosome could lead to failure to inactivate feminising genes or activate masculinising ones, resulting in same-sex attraction in males. It could not be tested at the time, but some evidence came from X chromosome methylation patterns in mothers of gay men.

Bocklandt later headed to the University of California, Los Angeles, and tested imprinting directly by comparing the entire genome’s epigenetic marks in male identical twins of whom one was gay and one straight. He looked at 30,000 methylation sites in all, but though several correlations turned up, this could not be repeated.

The latest study by Tuck Ngun, also at UCLA, scans 4 million potential methylation points (see “‘I quit’: concern over gay genetics“). Five were identified which, together, could classify the sexual orientation of 37 gay/straight twin pairs with 67 per cent accuracy. Ngun sent an abstract based on preliminary data to the American Society of Human Genetics, hoping for feedback at its conference last week.

The society issued its own press release about the work to major media, and not surprisingly there were soon headlines about the new “gay gene test” and concerns about abuses, such as elimination of homosexuality via abortion. Some of the coverage created misconceptions that need correcting. Firstly, Ngun’s work does not amount to a sexual orientation test. Even if it can be replicated in more twins with highly correlated methylation patterns, it is unlikely to work in unrelated members of the public.

What’s more, the study doesn’t discount the idea of genes directly influencing sexuality. There is room for genes and imprinting, and in fact they may synergise. That means epigenetics may only make a difference when combined with a certain genetic background.

Finally, such work won’t worsen homophobia. People who understand the role of biology in sexuality are more likely to be accepting and inclusive.

I hope Ngun’s findings will stimulate more research. My fear is that the furore stirred up will inhibit it. That would be a pity, because sexual orientation is one of the most fundamental and fascinating variations in humanity that we can study.

(Image: George Panagakis/Pacific Press/LightRocket via Getty)

This article appeared in print under the headline “Can a DNA test reveal sexuality?”

The idea of testing someone’s DNA to determine whether they are gay or straight is provocative. So it was no surprise that a recent study by Tuck Ngun at the University of California, Los Angeles, entitled ““, caused a stir.

Talk of a possible test is unhelpful. As a scientist long interested in the role of the genome in human sexuality, I am more intrigued by what the study can tell us about the influence of epigenetic imprinting – the silencing of genes by methylation of DNA. This chemical imprint is passed from parent to child and has implications for a variety of complex human traits. My lab proposed 12 years ago that it plays a role in sexuality.

To understand what all the fuss is about now, it’s essential to know some history. The first DNA markers for sexual orientation were , when we discovered an area on the X chromosome – Xq28 – that influences sexuality in gay brothers. This year, that finding was robustly , which also found an area on chromosome 8 that plays a role.

Despite these so-called linkage studies, it was clear that inherited differences in DNA could not account for all of the observed variation in sexual orientation. That led me and graduate student to hypothesise in 2003 that an .

X chromosome

We illustrated the idea by suggesting how atypical imprinting of the X chromosome could lead to failure to inactivate certain genes or activate others, thereby leading to same-sex attraction in males. Although we couldn’t test this directly at the time, some supporting evidence emerged when the of the mothers of gay men were compared with those seen in the mothers of straight men.

When Bocklandt left for a postdoctoral position with at UCLA, another author on Ngun’s study, he proposed testing imprinting directly by comparing the methylomes – the entire genome’s epigenetic marks – from pairs of male identical twins in which one is gay and one straight.

Despite scepticism from funders, Bocklandt and Vilain persisted, and began recruiting such pairs and analysing their DNA in a way that allowed comparison of some 30,000 methylation sites. Several intriguing correlations were found, but none of them held up to replication.

Ngun’s innovation was to expand the search to the entire genome using a new technology that covers a whopping 4 million potential methylation points. These were finally narrowed down to 140,000 points that were compared in 37 gay/straight twin pairs – a small sample for such a complex analysis.

Press release

Ultimately five sites were identified which, in combination, could be used to classify the sexual orientation of the twins with 67 per cent accuracy. Ngun submitted an abstract describing his preliminary data to the American Society of Human Genetics, hoping to get feedback and ideas for further testing from colleagues at its annual conference last week. This is a standard practice, and Ngun had presented the same data at previous meetings without fanfare.

The society’s press office then issued its own release about the work and circulated it to major media outlets. Not surprisingly, there were soon headlines about the new “gay gene test”. Some speculated that the research would lead to abuses, including the elimination of homosexuality through abortion.

Underlying the coverage were misconceptions about the research that need correcting. Firstly, it does not amount to a sexual orientation test. Even if the data can be replicated in more twins with highly correlated methylation patterns, the methodology is unlikely to work in unrelated members of the general population.

What’s more, the study doesn’t discount the previous linkage results. There is plenty of room for both genes and imprinting, and in fact they may synergise. That means the epigenetic differences, which are by themselves small, may make a difference only when combined with a particular genetic background.

Finally, such work will not lead to increased homophobia. Scientific knowledge is one of the best defences against the prejudice that underlies discrimination. As I noted in a , surveys show that people who understand the role of biology in sexuality are much more likely to be accepting and inclusive.

I hope that Ngun’s preliminary findings will stimulate more research. My fear is that the unnecessary furore stirred up by the press release will inhibit it. That would be a pity, because sexual orientation is one of the most fundamental and fascinating variations in humanity that we can study.

(Image credit: George Panagakis/Pacific Press/LightRocket via Getty)