EXPERT REACTION: There is no single 'gay gene'

Embargoed until: Publicly released:

There is no single “gay gene”, say Aussie and international researchers who analysed survey responses from people about their same-sex sexual behaviour, and analysed genetic data from over 470,000 people. The researchers could not find any way to meaningfully predict or identify a person’s sexual behaviour on the basis of their genes. While they did find five genetic variants associated with same-sex behaviour (and lots of others which may also be involved) at best these genetic differences could only account for between 8 and 25 per cent of variation in same-sex sexual behaviour and could not be used to predict it. The findings suggest same-sex sexual behaviour is influenced by a complex mix of genetic and environmental influences, similar to most other human traits.

Journal/conference: Science

DOI: 10.1126/science.aat7693

Organisation/s: The University of Queensland, Monash University, The University of Sydney, The University of Western Australia, University of South Australia

Funder: A.R.S. received funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development specifically to investigate the genetics of sexual orientation: R01HD041563 (A.R.S., principal investigator) and R21HD080410 (A.R.S. and E.R.M., multiple principal investigators). E.R.M., G.W.B., and S.G. are also supported by R21HD080410. No other member of the group received funding specifically for this study, but members of our team received salary funding from organizations as well as our own universities. B.P.Z. received funding from The Australian Research Council (FT160100298). A.G. was supported by the Knut and Alice Wallenberg Foundation (2015.0327) and the Swedish Research Council (2016-00250). A.G., R.M., and B.M.N. were supported by National Institutes of Health (NIH) grant 1R01MH107649-03 (to B.M.N.). R.W. was supported by the National Science Foundation’s Graduate Research Fellowship Program (DGE1144083). Any opinion, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. M.G.N. is supported by ZonMw grants 849200011 and 531003014 from the Netherlands Organisation for Health Research and Development. This research used data from Add Health, a program project directed by K.M.H. (principal investigator) and designed by J. R. Udry, P. S. Bearman, and K.M.H. at the University of North Carolina at Chapel Hill, and funded by grant P01-HD031921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. Information on how to obtain the Add Health data files is available on the Add Health website ( This research used Add Health GWAS data funded by Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) grants R01 HD073342 to K.M.H. (principal investigator) and R01 HD060726 to K.M.H., J. D. Boardman, and M. B. McQueen (multiple principal investigators). The genetic part of the CATSS study was supported by grant 2014-0834 from the Swedish Council for Working Life and Social Research.

Media Release

From: AAAS

According to a genome-wide association study involving more than 470,000 people, a person’s genetic variants do not meaningfully predict whether they will engage in same-sex sexual behavior. The findings suggest same-sex sexual behavior is influenced by a complex mix of genetic and environmental influences, similar to what’s seen for most other human traits. There is no single “gay gene,” the study’s authors say, and instead there are thousands of genetic variants linked to the trait, each with small effects. Andrea Ganna et al. examined the genetics of individuals who self-reported on whether they had ever engaged in same-sex sexual behavior. The authors analyzed survey responses and performed genome-wide association studies (GWAS) on data from over 470,000 people in the UK Biobank and 23andMe, Inc. The researchers could not find any patterns among genetic variants that could be used to meaningfully predict or identify a person’s sexual behavior, they say. “[M]any loci with individually small effects…additively contribute to individual differences in predisposition to same-sex sexual behavior,” they write, describing genetic patterns consistent with many personality, behavioral, and physical traits. In their study, only five genetic variants were “significantly” associated with same-sex behavior, and thousands more appear to also be involved, but taken together these variants had only small effects and are far from being predictive, the authors emphasize. They note that some among these variants are linked to the biological pathways for sex hormones and olfaction, providing clues into mechanisms influencing same-sex behavior. “Our findings provide insights into the biological underpinnings of same-sex sexual behavior,” say Ganna et al., “but [they] also underscore the importance of resisting simplistic conclusions because the behavioral phenotypes are complex, because our genetic insights are rudimentary, and because there is a long history of misusing genetic results for social purposes.”

In a Perspective, Melinda Mills emphasizes the limitations of the study results: “…although they did find particular genetic loci associated with same-sex behavior, when they combine the effects of these loci together into one comprehensive score, the effects are so small (under 1%) that this genetic score could not be reliably used to predict same-sex sexual behavior of an individual.” She adds that “using these results for prediction, intervention or a supposed ‘cure’ is wholly and unreservedly impossible.”

Expert Reaction

These comments have been collated by the Science Media Centre to provide a variety of expert perspectives on this issue. Feel free to use these quotes in your stories. Views expressed are the personal opinions of the experts named. They do not represent the views of the SMC or any other organisation unless specifically stated.

Associate Professor Craig Smith is a developmental biologist and geneticist working on sex development at Monash University.

Sex and gender are hot topics in society. Witness the same-sex marriage debate and the recent passing of legislation in Victoria allowing transgender people to self-nominate their sex on their birth certificates.

A research paper published today in the international journal, Science, visits the question of sexuality and genetics.

A long-standing issue in this area has been the old 'nature' versus 'nurture' debate. That is, to what extent is our sexuality moulded by our genes or by the environment?

The current study looked at the genetics side of this debate, studying over 477,000 people in the UK, focusing on same-sex sexuality. The researchers found no clear single gene implicated in same-sex sexual behaviour.  Nor did it resolve the nature versus nurture issue.

The researchers used a method called a Genome Wide Association Study (GWAS) which essentially compares genetic variation among people to see if a particular variation is linked to a trait, in this case same-sex behaviour.

They found five regions of the human genome (genes) that were statistically associated with same-sex behaviour. But overall, these regions could only account for up to 25 per cent of the variation seen in sexual preference. This genetic variation could not predict an individual’s sexual preference.

The researchers concluded that there are many genes, each with individually small effects – and partly overlapping in males and females- adding up to contribute to an individual’s predisposition to same-sex behaviour.

The significance of this work is that it is one of the largest and most statistically rigorous studies of its kind. While the findings are important for our understanding of human sexuality, we shouldn’t be terribly surprised by them. Complex human traits such as sex, sexuality and gender identity will involve many genes. There is very unlikely to be one gene governing any of these traits.

What about nurture?  The current study was focused on the genetics side and makes no conclusions about the potential role of 'environment'.  The authors state that their results 'overwhelmingly point towards the richness and diversity of human sexuality.'

Last updated: 20 Jan 2020 5:21pm
Declared conflicts of interest:
None declared.
Dr Traude Beilharz is a Senior Research Fellow in the Department of Biochemistry & Molecular Biology at Monash University

Taylor Swift tells us that ‘shade never made anyone less gay’ and neither will this research.

From Proust to 50 Shades we know that human desire is complicated. And for that we can be GLAAD! Diversity brings the colour to life on earth.

Genomic research helps us to understand how genetic information relates to wellness and disease.

However, in addition to the DNA sequence variation reported here, each human genome is decorated with a layer of epigenetic variation that reflects our environmental exposures.

Unlike DNA which is stable and nearly identical in each cell of our body, this epigenetic coat can differ between cell types and can change over time.

The complexity this brings means we are very far from understanding how genomic variance influences behaviour.

For now, to understand what it means to be gay, it might just be easier to have a conversation.

Last updated: 29 Aug 2019 4:33pm
Declared conflicts of interest:
None declared.
Fahad Ali is a Research Assistant at the School of Life and Environmental Sciences, the University of Sydney

Like all complex human traits, sexual orientation cannot be reduced to a genetic on/off switch, and the findings by Ganna and colleagues confirm this. They have identified five locations in the human genome where a single letter change in the genetic code (out of three billion) seems to be very slightly associated with the experience of ever having sexual activity with someone of the same sex. 

It should be understood that these genetic variations do not determine sexual orientation in any predictive way, and you can have all of these variants without being same-sex attracted (and vice versa).

Last updated: 29 Aug 2019 4:32pm
Declared conflicts of interest:
None declared.
Dr James Morandini is a Postdoctoral Fellow at Social Cognition Individual Differences Laboratory, School of Psychology, the University of Sydney

This is one of the largest studies to-date examining genetic contributions to same-sex sexual behaviour.

The study found there is no single ‘gay gene’, but rather there are multiple gene variants that each predict some small amount of the variance in whether someone has engaged in same-sex sexual encounters in the past. 

Complicating the picture further, the gene variants predicting same-sex sexual behaviour differed between males and females and differed depending on the proportion of same-sex to other-sex partners an individual had. 

These findings suggest that the genetic architecture of same-sex behaviour is complex and that there is no single genetic etiology for same-sex behaviour.

Last updated: 29 Aug 2019 4:31pm
Declared conflicts of interest:
None declared.
Associate Professor Greg Neely is Head of the Dr John and Anne Chong Laboratory for Functional Genomics, Charles Perkins Centre, the University of Sydney

This study used large-scale detection of DNA polymorphisms in about 26,000 people who report having had same-sex partners to identify areas in the human genome that correlate with sexual preference.

The researchers found from 8 per cent to 25 per cent of the factors influencing sexual preference arise from genetic contributions, although the specific parts of the genome identified in this study combined only contribute to 1 per cent of sexual preference and could never be used to predict sexual preference. 

While the study is based on correlation and requires further evidence, the data implicate the sense of smell, and sex determination during development, as factors that may influence sexual preference. 

A major weakness of this study is that it is primarily based on data from 40- to 70-year-old people across the UK, and same-sex preference may be underrepresented based on societal pressure from a previous era.

Another issue is that the researchers appear to link sexual preference with mental health issues. However, we already know that sexual minorities are more likely to suffer from depression or anxiety and this can easily be the result of social prejudice or isolation, so in my opinion this component of the study was not adequately justified or informative.

Importantly, this study shows that sexual preference is not a single dimension and same-sex attraction is not inversely related to opposite-sex attraction, arguing for a more nuanced social understanding of sexual orientation that also includes bisexuality and asexuality.

Last updated: 29 Aug 2019 4:30pm
Declared conflicts of interest:
None declared.
Dr Brendan Zietsch is an ARC Future Fellow in the School of Psychology at the University of Queensland

We have long known, from studies of twins, that a person’s sexual preference is influenced by his or her genetic makeup. What we have not known is what specific gene, or genes, are important.

Our international team drew together two huge samples, totalling nearly half a million participants whose DNA and sexual behaviour were known. We looked across the whole genome for locations where one letter of DNA code was more or less common in participants who reported having had same-sex sexual partners than in those who reported only opposite-sex partners. What we found was that there is no one “gay gene” – instead, there are many, many genes that influence a person’s likelihood of having had same-sex partners.

Individually, each of these genes has only a very small effect, but their combined effect is substantial. We could be statistically confident about five specific locations; we could also tell with high confidence that there are hundreds or thousands of other locations that also play a role, though we couldn’t pinpoint where they all are.

Follow-up analyses of significant genes suggested that sex hormones may be involved in sexual preference. Overall, our results reveal a highly complex genetic influence on same-sex sexual behaviour. Because of this complexity, we cannot meaningfully predict a given person’s sexual preference from their DNA – nor was this our aim.

Moreover, it is important to note that sexual preference is influenced by genes but not determined by genes. Non-genetic influences are also important, but we know little about these and our study does not shed light on them.

Last updated: 29 Aug 2019 4:29pm
Declared conflicts of interest:
None declared.
Professor Bob Williamson is the Leader of the Machine Learning Group at NICTA.

No one will be surprised that sexual attraction is in part determined by genes and in part by environment. This is true for every form of human behaviour.

Some individuals who are gay or lesbian will be more influenced by the genetics side, some will be more influenced by environmental factors. We know that the proportion of people who are same-sex attracted varies from time to time, and from country to country, much too quickly to be 'only genetic'.

This study of almost half a million individuals (who self-reported their sexual preferences) had the great advantage of looking at a very large number of people, but this also means that both genes and environment are 'averaged' over the population – more careful study may reveal sub-groups where a particular gene, or environmental trigger, is particularly important.

Several hundred genes appear to have an influence, so the researchers conclude that there is no 'gay gene', and their data cannot be used to predict if someone is gay or straight.  

The study found that gay men and lesbian women share about half the same gene predictors, but not all.  And most of the samples were from Britain (from the public UK Biobank), but the results agreed with those from 23andMe, a very large US ancestry and health company which had DNA stored from many people who identified as gay.

This agreement gives the researchers more confidence – when it comes to genes, two populations are better than one, and the results will almost certainly be similar for Australia.

These genome-wide studies are great for giving leads for future research, but cannot be used to tell us about individuals. Scientists who work in controversial fields such as this must speak up, advocate, and ensure that their findings are not used, falsely, to discriminate against anyone who is gay or lesbian. 

As Melinda Mills emphasised in an accompanying perspective, 'using these results for prediction, intervention or a supposed ‘cure’ is wholly and unreservedly impossible.'

Last updated: 29 Aug 2019 4:26pm
Declared conflicts of interest:
None declared.
Professor Madeleine Beekman is the Deputy Head of School and Chair of Ecology, Evolution and Environment at the School of Life and Environmental Sciences, University of Sydney

Francis Galton, half-cousin of Charles Darwin, was probably the first to attempt to find evidence for the inheritance of human traits. Heights initially, but later also intelligence.

When plotting height against the number of people of a particular height, Galton found a bell-shaped curve.

Galton was convinced that height was heritable, but it would take many more decades before scientists were able to find evidence for the heritability of height in humans.

In 2017, more than a decade after the first genome-wide association study on humans, a consortium of scientists, using a data set from more than 700,000 people, showed that the total number of genes that influence height is almost eight hundred.

Human height is a quantitative trait; a trait influenced by many genes and the environment. Such traits, when plotted against frequency, often show the bell-shaped curve Galton first described.

Now it appears that sexual preference is also a quantitative trait. Instead of a single gene, or a small number of genes, Ganna and colleagues found that many genes influence human sexuality.

But even these genes only accounted for a small amount of variation in human sexuality. The new study illustrates how the notion of a ‘gene for’ is a gross simplification.

Yes, there are a few instances where a single gene has a huge effect, but mostly complex traits are just that; complex.

Sexuality can now be added to the list of complex human traits."


ADDITIONAL COMMENT SENT VIA USYD MEDIA TEAM: "As with most human traits, this study finds that sexual behaviour is influenced by many genes. Hence, there is no single gene that influences sexuality in humans.

But even the genes that were found to be associated with sexual behaviour only explained a small fraction of the variation. Sexual behaviour is clearly the result of genetic and environmental factors.

Last updated: 29 Aug 2019 4:24pm
Declared conflicts of interest:
None declared.
Dr Nina McCarthy is a Research Fellow in the Centre for Genetic Origins of Health and Disease (GOHaD) at the University of Western Australia

This paper assessed the genetic contribution to same-sex sexual behaviour in 477,522 individuals from the United Kingdom and the United States, using information from genetic variants spread across the genome.

This is a much larger study than previous Genome Wide Association Studies (GWAS) of sexual behaviour, and therefore we can be more confident in the accuracy of the findings. 

This paper found that in their study sample, human sexuality is substantially influenced by genetic factors (in keeping with previous reports from analysis of relatives).

Like for many other human traits, many thousands of different genetic variants spread across the genome each make a tiny contribution to the overall genetic influence on sexuality. In general, how well we can predict a trait is inversely proportional to how many genes are involved (with good predictions for traits caused by only a few genes, such as eye colour).

It is not surprising therefore that collectively, the genetic variants studied in this paper were able to predict less than 1 per cent of the variance in sexuality in other samples, and could not be used to predict sexual behaviour on an individual level. 

Although their main analysis involved comparing heterosexual to non-heterosexual people, the authors also examined some other aspects of sexual orientation and behaviour.

They found that somewhat different genes influenced same-sex behaviour in females vs males. They also found that the genetic contribution to the proportion of same-sex partners and engagement in same-sex sexual behaviour appears to be different. 

The authors suggest some practical applications of their research. For example, they point out that their findings suggest some of the tools commonly used in research of sexual orientation may be based on misconceptions about the underlying structure of sexual orientation, and need to be rethought.

They also highlight the limitations of their research, a major one being that their study sample was not very diverse, including only participants of European ancestry and largely from the US and the UK – ‘research involving larger and more diverse samples will afford greater insight into how these findings fare across different sociocultural contexts’.

In addition, the study sample did not include transgender persons, intersex persons, and other important persons and groups within the queer community, a limitation which the authors say they hope will be addressed in future work.

Overall, I think that the results of this paper indicate that the genetic underpinnings of human sexuality are extremely complex, diverse and nuanced. I believe that well-conducted genetic studies into human sexuality such as this may help to inform, and perhaps have a positive impact on, our understanding of human sexual diversity.

Last updated: 29 Aug 2019 4:22pm
Declared conflicts of interest:
None declared.
Dr Sang Hong Lee is a senior lecturer in statistical genetics at the Centre for Population Health Research, School of Health Science at University of South Australia

The authors did a great job to dissect the genetic architecture of same-sex behaviour as a complex trait, by using a number of statistical genetics methods, e.g. GWAS, estimating single nucleotide polymorphism (SNP)-based heritability and genetic correlation analyses.

This study is important and clarifies a number of crucial points, using a large-scale biobank data, that is 1) the same-sex behaviour is a polygenic trait, 2) a substantial proportion of phenotypic variance is explained by genome-wide SNPs, 3) the trait is genetically associated with a wide range of other traits such as mental health, personality and reproductive traits, 4) same-sex behaviour may not be the same trait between males and females (i.e. across-sex genetic correlation is different from 1).

Although the authors reported that the predictive power is limited (less than 1 per cent variance explained by polygenic risk scores), it can potentially increase, depending on the reference sample size, if the reported total variance (8 to 25 per cent) is genuine.

I expect there will be a number of following studies to confirm their findings and to reveal more insights into the trait. It is also crucial these findings can be replicated in a larger random sample (note that UK Biobank and 23andMe are not random samples).

Finally, I agree with the authors underscoring the importance of resisting simplistic conclusions.

Last updated: 29 Aug 2019 4:19pm
Declared conflicts of interest:
None declared.

News for:


Media contact details for this story are only visible to registered journalists.