New Paper Proposes Epigenetic Pathway to Homosexuality (UPDATED)

I am still reading the paper by William Rice, Urban Friberg and Sergey Gavrilets (available for review here) and have asked several other scientists for their opinion of it.  It is a complex argument but one which initially seems compelling to me.

Here is the abstract:

Male and female homosexuality have substantial prevalence in humans. Pedigree and twin studies indicate that homosexuality has substantial heritability in both sexes, yet concordance between identical twins is low and molecular studies have failed to find associated DNA markers. This paradoxical pattern calls for an explanation. We use published data on fetal androgen signaling and gene regulation via nongenetic changes in DNA packaging (epigenetics) to develop a new model for homosexuality. It is well established that fetal androgen signaling strongly influences sexual development. We show that an unappreciated feature of this process is reduced androgen sensitivity in XX fetuses and enhanced sensitivity in XY fetuses, and that this difference is most feasibly caused by numerous sex-specific epigenetic modifications (“epi-marks”) originating in embryonic stem cells. These epi-marks buffer XX fetuses from masculinization due to excess fetal androgen exposure and similarly buffer XY fetuses from androgen underexposure. Extant data indicates that individual epi-marks influence some but not other sexually dimorphic traits, vary in strength across individuals, and are produced during ontogeny and erased between generations. Those that escape erasure will steer development of the sexual phenotypes they influence in a gonad-discordant direction in opposite sex offspring, mosaically feminizing XY offspring and masculinizing XX offspring. Such sex-specific epi-marks are sexually antagonistic (SA-epi-marks) because they canalize sexual development in the parent that produced them, but contribute to gonad-trait discordances in opposite-sex offspring when unerased. In this model, homosexuality occurs when stronger-than-average SA-epi-marks (influencing sexual preference) from an opposite-sex parent escape erasure and are then paired with a weaker-than average de novo sex-specific epi-marks produced in opposite-sex offspring. Our model predicts that homosexuality is part of a wider phenomenon in which recently evolved androgen-influenced traits commonly display gonad-trait discordances at substantial frequency, and that the molecular feature underlying most homosexuality is not DNA polymorphism(s), but epi-marks that evolved to canalize sexual dimorphic development that sometimes carryover across generations and contribute to gonad trait discordances in opposite-sex descendants.

The Christian Post did a quick piece on this study but did not interview anybody with any expertise about the research. Instead, they reached out to David Pruden who made the following statement:

“The theoretical model itself attributes only 10-14 percent of the factors to genetics or epigenetics. That leaves the remaining 85 percent or so of the factors to environmental influences,” said Pruden.

I don’t see anything like that in the paper so I can’t evaluate what he is claiming here. This sounds like the usual NARTH misreading of the genetic evidence but not the Rice et al paper.

UPDATE: Dr. Rice wrote to say that Pruden’s account is inaccurate. In response to my question: “Are you intending to account for homosexuality whenever it occurs?” Dr. Rice replied:  “Yes, we think that carry-over epi-marks have the potential to account for most homosexuality.” 

More to come…

"Slapping the word "conscientious" on objection to war service DID exempt "conscientious objectors" from the ..."

Sponsor: Celibacy as Therapy Goal Allowed ..."
""Activity A includes Activity B" means that if A is forbidden, B is forbidden as ..."

Sponsor: Celibacy as Therapy Goal Allowed ..."
"White privilege is something that exists equally for the most pius, and saintly among us ..."

Is Colorblind Theology Blind?
"Wash your hands before every meal.Brush and floss at least twice a day.Don't post non ..."

Can You Trust Glenn Beck’s Mercury ..."

Browse Our Archives

Follow Us!

What Are Your Thoughts?leave a comment
  • Richard Willmer

    The ‘correction’ you mention in your UPDATE, Warren, has now been appended to the CP article itself.

  • Lynn David

    Wonder if it will in any way tie into the incomplete methylation of the X-chromosome in males (what I was calling strong-X 6 or so years ago).

    I looked through it and did not see percentages such as Pruden mentioned as well.;

  • Lynn David

    Well, that’s what happens when I write something and don’t hit send for over an hour! Good to hear Dr Rice’s explanation.

  • Ann

    Dr. Throckmorton,

    Have you ever considered egigenetics before this paper? I remember sending you some articles on it before to consider but didn’t get a response.

  • Lynn David

    Ann… Epigenetics has been discussed here at least for 5 years. The work by UCLA and the work they were attempting at Chicago had to do with epigenetic causes, such as, the incomplete shut down (methylation) of the X chromosome in the case of male homosexuality.

  • Ann

    Thank you Lynn David – I don’t remember the term “epigenetics” being used except when I wrote about it. I guess I didn’t realize the connection you mention and do now.

  • Karen


    Will you be posting something at some point in layperson terms? Also, does this adequately account for women as well as men? So many studies seem skewed toward men. It would seem that since men’s and women’s sexualities are rather different, that a one size fits all theory would be problematic. Also, is there any way to actually test this in persons? For example, say this is true, would I then be able to go in for a test that shows I have these markers which make me gay?

  • carole

    Lynn David,

    What hear you about the UCLA epigenetic study?

  • Lynn David

    Nothing, Carole.

  • carole


    Are you familiar with the private company, 23andme? It’s a biotech and genomics company. People send in their dollars and their DNA and the company provides them with genetic info.

    Sometimes they ask their clientele what study would most interest them. Turns out that one of the most popular areas of interest was sexual orientation so the company has an on-going GWAS of sexual orientation. They recently presented their results at the American Society of Human Genetics meeting in San Francisco last month.

    general link to study’s background:

    Here are the study’s results up to this point.

    Key points:

    They didn’t find as of yet (after approx. 23,000 samples) any genetic loci reaching genomewide significance “nor any evidence of an association between sexual identity and SNPs on the X chromosome” in men and women. They are continuing to collect samples.

    Their results replicated phenotypic characteristics of other studies concerning addiction, mental health issues.

    Also interesting since even a non-significant “hit” can lead to clues:

    “Interestingly, our top (non-significant) hit was in 8q12.3 for men, which was also identified as a top (non-significant) hit in a previous

    linkage study among a small sample of gay men using

    microsatellites (2) as well as a current ASHG abstract (9).”

    I looked, therefore, at note 9 and saw that it was the abstract from what looks to be several years’ long gay brothers study from Sanders & Bailey, et al: “Sanders AR, Dawood K, Rieger G, Badner JA, Gershon ES, Krishnappa

    RS, Kolundzija AB,Guo S, Beecham GW, Martin ER, Bailey JM. Genomewide

    linkage scan of male sexual orientation. American Society of Human

    Genetics 2012 conference presentation 1957T.”

    All I can find from that is the abstract:

    I don’t know much about the numbers and how significant or insignifcant these numbers from the Sanders study are or aren’t. Have been waiting to read the study but haven’t found it yet.

    I’d expect to see the full study pretty shortly.

  • Gregory Peterson

    Don’t forget micro-RNA in gene regulation…though if that could have anything to do with expressions of sexuality, I wouldn’t know. I’m an artist.

    I vaguely wonder if epigenetics could explain the difficult to diagnose expression of my pernicious anemia. Talk about a kick in the head… At least I’m not going weirdly blind anymore.

  • Zoe Brain

    Warren, we really, really need to have a talk about Gender Identity. Maybe offline.

    Here’s what I’ve been saying for a while:

    Sexual Hormones and the Brain: An Essential Alliance for Sexual Identity and Sexual Orientation Garcia-Falgueras A, Swaab DF Endocr Dev. 2010;17:22-35

    The fetal brain develops during the intrauterine period in the male direction through a direct action of testosterone on the developing nerve cells, or in the female direction through the absence of this hormone surge. In this way, our gender identity (the conviction of belonging to the male or female gender) and sexual orientation are programmed or organized into our brain structures when we are still in the womb. However, since sexual differentiation of the genitals takes place in the first two months of pregnancy and sexual differentiation of the brain starts in the second half of pregnancy, these two processes can be influenced independently, which may result in extreme cases in trans-sexuality. This also means that in the event of ambiguous sex at birth, the degree of masculinization of the genitals may not reflect the degree of masculinization of the brain. There is no indication that social environment after birth has an effect on gender identity or sexual orientation.

  • Richard Willmer

    @ Zoe

    I reckon you were always right about that. And this paper does very seem to point very much in your direction. 🙂

  • Warren Throckmorton

    Zoe – In the past 2-3 years, don’t think I have disagreed with that general approach have I? For quite some time, I have been skeptical that social environment after birth has much effect on GI or SO. Perhaps our difference is that I think some effect might be a safer way of accounting for the evidence whereas I think you are suggesting there is no effect?

  • Richard Willmer

    Something that probably does need a ‘closer look’ is the relationship between GI and SO. One of my closest friends is a transgendered woman and the sense from my many conversation with her is that they really are almost entirely separate matters. There’s also a problem with ‘categories’: she’s a transgendered woman who is attracted to women. What does that make her? Heterosexual? Lesbian? What?

    Perhaps we should ‘take seriously’ GI, before seeking to ‘categorize’ SO? I don’t know (I’m just ‘thinking aloud’).

  • Ann

    If this has already been discussed, please be patient with my question.

    One of my sisters-in law is a twin – her orientation is completely straight and her fraternal twin sister’s orientation is gay. Since they have the same father who could have passed on the epi mark that was not erased, wouldn’t both babies be affected in the womb or would that be applied to identical twins only?

  • Zoe Brain

    Warren – No effect? Theory says there *have* to be borderline cases, neuro anatomy and biology are not binary, and a weak gender identity can crystallise in either category, or fluctuate without crystallisation.

    If there are no genderfluid or genderqueer people, our theory is wrong. But they have also to be a minority. In such cases, environment has to play a significant role there, they’re just the exception, not the rule.

    So yes, there’s an effect, there has to be, in some cases. But mostly not.

    Look at the results of surgery on unconsenting hormonally-anomalous Intersex kids. If there was zero environmental effect, we’d have a 50% success rate by assigning an arbitrary gender.

    Instead, catatrophic failure rate is closer to 30% than 50%. That implies that when an Intersex condition involves anomalous hormones in the womb, many are close enough to the borderline to adapt as either sex, though possibly not adapt terribly completely.

    On hormonally usual Intersex kids, we’d have close to 100% success (if the congruent gender) or 100% failure (if incongruent). As we do by assigning hormonally-usual XY children with cloacal extrophy as female), that’s nearly always a disaster. See also the David Reimer case, In such circumstances, the degree of “bigender” appears much the same as for the general population, though whether that’s 5%, 10% or 20% is not well known.

    Remember, gender (we think) only comes to be determined in the context of environment. If a child is confronted with a group that has similar emotional responses (I think that’s the greatest factor) , instincts, thought patterns, similar sense of smell and hearing, then they will come to identify as a member of that group, even if they look and dress differently..

    Biased-Interaction Theory of Psychosexual Development: “How Does One Know if One is Male or Female?” M.Diamond Sex Roles (2006) 55:589–600

    A theory of gender development is presented that incorporates early biological factors that organize predispositions in temperament and attitudes. With activation of these factors a person interacts in society and comes to identify as male or female. The predispositions establish preferences and aversions the growing child compares with those of others. All individuals compare themselves with others deciding who they are like (same) and with whom are they different. These experiences and interpretations can then be said to determine how one comes to identify as male or female, man or woman. In retrospect, one can say the person has a gendered brain since it is the brain that structures the individual’s basic personality; first with inherent tendencies then with interactions coming from experience.

  • Zoe Brain

    Richard Willmer –

    Something that probably does need a ‘closer look’ is the relationship between GI and SO. One of my closest friends is a transgendered woman and the sense from my many conversation with her is that they really are almost entirely separate matters. There’s also a problem with ‘categories’: she’s a transgendered woman who is attracted to women. What does that make her? Heterosexual? Lesbian? What?

    If GI and SO were the same, Lesbian Transwomen and Gay Transmen would not exist. They do, in significant numbers. Therefore GI and SO are different.

    Terminology is a problem, a big one. I regard anyone not floridly psychotic who self-identifies as female as a woman, regardless of other circumstances.

    That’s regardless of whether she was :

    The “standard factory model” – XX chromosomes, female genitalia, assigned female at birth.

    An XY woman :XY chromosomes, female genitalia, assigned female at birth.

    An Intersex XX woman: XX chromosomes, ambiguous genitalia, assigned male at birth

    A Trans woman : Usually XY, sometimes XXY, rarely XX, male or mostly male genitalia, assigned male at birth;

    And many, many other possible combinations.

    Otherwise we have absurdities such as “transvestite homosexual men” who are gynephillic and do not cross-dress, just because they identify as women – and have neuro-anatomy, and often other anatomy, to match.

    If GI and SO were wholly unconnected, due to societal pressures, then the number of Gay and Lesbian Trans people would reasonably be expected to be larger than the number of Straight ones. They’re not. So there is some correlation, just a weak one.

    If we posit that different parts of the brain determine GI and SO, and that feminisation or mascuinisation of GI due to hormonal environment in the womb strongly biases the odds of the same feminisation or masculinisation of SO, but that feminisation/masculinisation of SO only weakly biases the odds of feminisation/masculinisation of GI, then we get similar numbers to what is observed.

    We’d expect to observe both Stone Butch lesbians, and Lipstick lesbians, Twinks and Bears, in both cisgendered (ie not transgendered) and transgendered populations.

    The difficulties with determining SO are twofold:

    1) Nearly everyone’s SO changes. It goes from Asexual to Androphillic, Gynephillic, or both to a greater or lesser degree, at puberty. So what is established before birth is a potential, apparently an unchangeable one, that is activated by a hormonal flood at puberty.

    GI on the other hand, though it might crystallise late or not at all in boundary cases, is usually evidenced long before puberty, as early as age 2 in unusually strongly gendered cases, usually around age 5-6, nearly always by age 10.

    An “effeminate boy” is likely to be Androphillic after puberty, but Gender Identity may well crystallise as male. A “tomboy” has a slightly higher chance of being Gynephillic, but Gender Identity is likely to crystallise as female.That’s regardless of whether either are brought up as male or female, that will only have an effect in boundary cases.

    2) The other problem is how do we determine if others are male or female? Is our SO dependant on pheremones, appearance, observation of behaviour and body language, socially-determined cues such as dress?

    I don’t consider a man who is attracted to another male that is a very convincing “drag queen”, mimicking female mannerisms, dress, behaviour etc to be Androphillic.

    Here there are obviously instinctive, hard-wired attractions at work, but equally obvious social factors – high heels, makeup, long hair that are associated with our concepts of sex.

  • Warren Throckmorton

    @Karen – Yes, hope to next week and if not then after the holidays…

  • Richard Willmer

    @ Warren

    You have been properly ‘cited’ in the ‘correction’. Good.

  • Zoe Brain

    A quick citation:

    Using the nationally representative sample of about 15,000 Add Health respondents in Wave III, the hypothesis is tested that masculinity–femininity in adolescence is correlated with sexual orientation 5 years later and 6 years later: that is, that for adolescent males in 1995 and again in 1996, more feminine males have a higher probability of self-identifying as homosexuals in 2001–02. It is predicted that for adolescent females in 1995 and 1996, more masculine females have a higher probability of self-identifying as homosexuals in 2001–02. Masculinity–femininity is measured by the classical method used by Terman & Miles. For both time periods, the hypothesis was strongly confirmed for males: the more feminine males had several times the probability of being attracted to same-sex partners, several times the probability of having same-sex partners, and several times the probability of self-identifying as homosexuals, compared with more masculine males. For females, no relationship was found at either time period between masculinity and sex of preference. The biological mechanism underlying homosexuality may be different for males and females.


    Journal of Biosocial Science / Volume 38 / Issue 06 / November 2006, pp 797-809

  • Teresa

    @Zoe, et. al.,

    OK, so here’s the future dilemma: the scientific community, in the not too distant future (less than a decade, I’d speculate) can identify SO through amniocentesis. Will parents opt for abortion rather than have a gay child, an inter-sex child? Will pro-life people be as pro-life as they claim?

    Downs Syndrome children (trisomy 21 and that whole family) are fast disappearing because of results of amniocentesis. Ethically, what’s the future for gay and inter-sex persons, if we can make them disappear? Will the Religious Right, many of whom consider homosexuals as responsible for most of society’s ills, regardless of behavior … will they own their child, full-well knowing their child will be gay, if that child chooses to be in a same sex relationship?

    Bioethics never catches up with science. I fear we have no idea where intrauterine discoveries will lead us.

    @Zoe, what’s your take on the ethics of this? Just because we can, do we?

  • Ann

    Will parents opt for abortion rather than have a gay child, an inter-sex child? Will pro-life people be as pro-life as they claim?


    I thought the same thing. Also, will advocates for a woman’s right to choose be as supportive when women choose to terminate the life of a baby who is pre-disposed to have a same sex orientation?

  • carole

    I like what Rice says about his model:

    “Rice’s model still needs to be tested on real-life parent-offspring pairs, but he says this epigenetic link makes more sense than any other explanation, and that his team has mapped out a way for other scientists to test their work.

    ” ‘We’ve found a story that looks really good,,’ he says. ‘There’s more verification needed, but we point out how we can easily do epigenetic profiles genome-wide. We predict where the epi-marks occur, we just need other studies to look at it empirically. This can be tested and proven within six months. It’s easy to test. If it’s a bad idea, we can throw it away in short order.’ ”

    Now we’ll see if there are any takers out there . Hey, 6 months? Why not?

  • Lynn David

    23andMe has my DNA. I did it mostly for genealogical reasons and I’ve found several cousins using DNA. But I have allowed my sample to be used in their studies.

  • carole

    “23andMe has my DNA. I did it mostly for genealogical reasons and I’ve found several cousins using DNA. But I have allowed my sample to be used in their studies.”