There are no genetic markers associated with sexual orientation

Great Debates
1

I have submitted my DNA to the 23andme organization a few years ago, and also (for the record) I identify as 100% homosexual (male).

According to their website, in researching 23,000 customers who voluntarily participated in research, they found no genetic markers associated with sexual orientation. Please note that the sexual orientations are all self-reported, and that these percentages are 89% mostly or exclusively heterosexual, 9% mostly or exclusively homosexual, and 2% bisexual. Unfortunately I cannot link everyone to the data because you have to be a member of 23andme to see it; if you are a member, here is the page (you may need to scroll down to see the data).

Bearing in mind the limitations of this research (voluntary participation, self-reported sexual orientation, possibly others) what does this do to the idea that sexual orientation is to some extent “built in” and not a choice, if we assume that the research is accurate?

In my opinion, this would switch the emphasis from genetic causes to such things as conditions in the womb during gestation. This could explain such observed phenomena as that later male children in large families are more likely to be homosexual, because although the genetic contributions of the parents would only change randomly from one child to the next, the conditions of the womb could be more likely to change in a predictable way from one child to the next. So, for example, a gestating child who is exposed to a very different mixture of sex hormones before birth compared to another child may have a different percentage expectation of a same-sex orientation.

In another way this finding is sort of a relief in that the likelihood of being able to purposely manipulate the sexual orientation of offspring by genetic manipulation using such tools as CRISPR, and thereby to reduce the percentage of same-sex orientation in the world, seems greatly reduced.

How do you view this data? Is it irrelevant or meaningful? If meaningful, in what way?

2

Last I heard, there was evidence for a genetic component to sexual orientation from twin studies: Identical twins have a higher correlation of sexual orientation than fraternal twins. Of course, this is not inconsistent with 23andme’s finding (or rather, lack of finding): A lot of genetics is very complicated, and everything involving psychology is very complicated. It’s quite possible that there is some genetic component to sexual orientation, but that it involves a large number of genes interacting in a manner too subtle to show up in their statistical tests.

That said, it’s also known that sexual orientation is not completely genetic: There is evidence that the prenatal uterine environment is also relevant, as you mention, and also evidence that upbringing is relevant. And even genetics, uterine environment, and upbringing all combined are still not enough to completely determine orientation, since there isn’t even perfect correlation between identical twins raised together.

3

Research to date always seemed to suggest a genetic component and I have no idea why I should trust 23andme as a research facility. So for now I’ll say “interesting. Wonder how this’ll shake out”.

4

That link goes to motion sickness for me but no matter. You have to put their negative finding in context of what they do.

23andme does not do complete DNA analyses; they have a custom designed chip that looks at the presence of absence of select single nucleotide polymorphisms (SNPs) that have been chosen as of interest by their researchers. These are a small fraction of all SNPs. So none of *the SNPs they look at *correlate with self-reported sexual orientation. Do others? Probably.

5

I would also say, not to be mean, but your relief is premature. People are going to continue searching for the reason behind homosexuality. It’s too fascinating to just forget about. So I don’t see much difference between discovering an in utero hormone therapy to stop it compared to a genetic test/therapy.

6

With all due respect, I’d suggest you publish your data in a peer review journal and then we can judge it.

Firstly, how are they defining “genetic marker”? That could be a single nucleotide or base pair or a long string of nucleotide pairs. They do not, to my knowledge, do a full genome analysis and so are generally looking at things above the level of individual genes.

Secondly, I don’t believe this is how genetic research even works. If you find something, good for you. But if you don’t find something, it might mean it doesn’t exist or you just didn’t look for the right thing or in the right place. Otherwise, we would have done this sort of thing a million times and found the genetic “markers” for all sorts of behaviors or maladies that we know are genetic or have a genetic component.

But keep in mind that this would be a huge news events. If scientist were to declare: It is now the scientific consensus there there is no genetic factor to homosexuality, you’d be reading about on the front page of the NYT and every news magazine out there.

7

I think 23andme has it wrong. There have been many twin studies showing a genetic link, even though studies don’t pinpoint which genes/alleles have an impact.

It’s complex and involves environmental as well as genetic factors, so there’s never going to be a recessive or dominant gay gene.

A Swedish study had about 1/3rd of the heritability of the trait in males being genetic, with less than 20% for females. Somehow I suspect a completely different set of genes are involved. Link: Biology and sexual orientation - Wikipedia

8

I don’t want to join that site just to see the study. Can you, without violating the copyright rules of this MB, quote the conclusion paragraph(s) of the study so we can see it? It’s really difficult to comment on something one hasn’t read.

9

This might be what he is talking about, which I think does not require membership.

About Genome-Wide Association Analysis (GWAS).

There are roughly 10 million SNPs in the human genome. Of those 10 million 23andme’s current custom chip read (v4) reads 602,000 total SNPs (the article’s listed “between 500,000 and 1 million”)

So the genetic contribution to sexual orientation is not one single one of those less than one out of ten all possible SNPs acting alone at least to the power as described in the article to detect. Not a finding without meaning but of limited meaning.

10

Wait, do I understand SNP to mean CG|AT|GC|TA ? Sure, there may be 10 million of those, but a bunch of them are things like the TATA markers and maybe some other non-transcribing sequences.

Which is not to defend their research. Just looking at the X/Y pair sounds very naïve to me. Why would we assume that preference is proximal to the genitals – when even the genitals do not always decode correctly?

Still, I personally doubt a genetic origin. Otherwise, sexual preference would track through families. Does it do that, or is it distributed evenly across the species? I mean, identical twins is a weak argument, because both of them experience the same uterine environment together anyway (if that is where it were to originate).

11

Twin studies are widely accepted as one of the best ways to test for genetic linkages. Identical twins are studied with fraternal twins as a control group. Scientists are not so naive as to forget to try and account for things like uterine environment as much as possible.

12

A question that occurs to me is what is the threshold for gayness? In other words, can you be a little gay and behave in a fully-fledged homosexual way, or can you feel very gay yet choose not to behave in a gay way at all? What I am really getting at is does everyone have some potential to become homosexual and if so at what point can we ‘define’ someone of either gender as being gay?

How do we know, for example, if it is not really a question of how much an individual is prepared to acknowledge homosexual feelings they experience and how much each individual wishes to pursue such feelings? Isn’t it the case that many people, at one time or another, feels attracted to some members of the same sex yet dismisses such ideas because of social norms? So, the question is how much does a conscious decision to ‘come out’ and choose to be a homosexual contribute towards being identified as gay? We know that public schoolboys who are subjected to long periods isolated from their families in boarding schools will be exposed to same-sex relationships in the absence of any affection from their natural families and perhaps will adopt this sexual orientation for life. So, I do not think it is always clear-cut as to what makes an individual gay or and fetal development, hormonal development and social attitudes all have an input but it is still far from clear what determines sexual orientation.

13

It’s important to note that genetic variants can be associated with some phenotype without directly contributing to its development. Any causal variant will be genetically linked with many non-causal variants on nearby regions of the chromosome. Of course, the strength of association between non-causal variants and a phenotype will be a function of how close they are to the causal variant. Thus, GWAS studies don’t need to test every variant, just enough to sample the entire genome with reasonable density.

When there are genetic loci with strong causal links to a phenotype, it’s possible to find statistically associated variants on the neighboring 10-20% of the chromosome. Thus, you only need to examine a few dozen SNPs per chromosome (for a total of a few hundred SNPs) to find associated regions.

In cases where there are a ton of genes with weak causal contributions to a phenotype, a well-powered GWAS study will need to examine higher density of variants, in many more individuals. With a simple trait like “height”, the most informative GWAS studies used 2.5 million SNPs and 250,000 individuals to find a few hundred variants that were collectively associated with 20% of height variation. In that example, “strong” variants were associated with less that 0.1% of height variation.

In comparison, I’m not surprised that this 23andme didn’t find an association. Compared to the height GWAS study, this sexual orientation GWAS study had a tiny fraction of the statistical power, and was examining a much more complex trait. It really just says that we can rule out the existence of a handful of causal variants which determine sexual orientation. But that’s the case for all complex traits, really: they’re produced by many subtle interacting effects from dozens or thousands of genes.

14

If we are able to pinpoint a “gay” gene, I could see a lot of parents wanting to eliminate it from their offspring’s germ lines. But I could also see some parents wanting to add it, for whatever reason. I’m curious if this scenario is bothersome to you as well.

At any rate, as long as a cause is identified–it doesn’t matter if it is genetic or environmental–parents are going to want to avoid it. Artificial wombs will one day be used to create “ideal” prenatal conditions for whatever kind of kid you want. And unlike with gene manipulation, where the parents have to be bold enough to explicitly request that a certain gene be eliminated, the guilt burden is lower with an artificial womb since ostensibly parents would only be seeking “ideal” prenatal conditions–one free from the ravages of air pollutants, poor nutrition, and maternal stress hormones and hormonal imbalances. Who wouldn’t want these things for their baby? I can see lots of people thinking that if the end result just happens to be a lower frequency of LGBTQ individuals in the population but the population as a whole is physically and mentally healthier, then maybe that’s an acceptable downside.

I hope this isn’t too much of a hijack, but I also wonder if transgendered people (in general) tend to take the same stand as homosexuals do (in general) wrt to the politics of gene selection. I can understand why a gay person would want to preserve “gayness”. Do transgendered individuals tend to see transgenderism in the same light, or do they view transgenderism more as a medical condition with root causes that should be mitigated or avoided, if possible?

15

The idea that sexual orientation (which itself is probably a misnomer) is due at least IN PART to womb conditions, is interesting. However, since the gist of the argument being made for it consists basically of “we didn’t see any genes wearing “I am gay” teeshirts, therefore it must be environmental,” it isn’t compelling.

After all, we have not found individual genes controlling all sorts of elements, which are clearly inherited.

One suggestion I have for everyone to think a bit about at least, is that the LABELS which we tend to limit ourselves to about this, are getting in the way of understanding what is really going on.

We have heterosexual, bisexual, homosexual, and asexual. But if we stop thinking of sexuality as being a simple yes or no proposition, it’s clear that there are all sorts of variations behind those labels. This suggests to me, at least, that sexuality is the result of collective affects of thousands of genetic, environmental, psychological, sociological, and even temporal factors.

Small examples: senses of smell, taste, hearing and so on, have been shown to play a part in attraction and desire. Genetics affects the mechanical ability of each of the senses to function.

We have seen that double-y chromosomes are correlated with some VERY nasty sexual behavior.

The existence of synesthesia, and the fact that it also has no known specific cause, AND has varying degrees of effect, points to the possibility that some aspects of sexual expression are the result of whatever it is which causes synesthesia.

There are lots of people in the world who “act gay,” in one way and another, but have zero interest in homosexual acts, and are only “turned on” by the opposite sex.

More than anything else, the basic fact that ALL of the studies of any kind, conducted so far, end up describing any and all findings, not as certainties, but as percentages of correlation, means that NONE of them by themselves, is the answer. Are children of older women slightly more likely to be gay, because of changes in the womb environment? Or is it because the eggs are older? Or even because “gay eggs” are less “aggressive” about making their way down the Fallopian tubes?

That last is me being silly, in case it isn’t obvious.

16

Identical twins share the womb, so how do twin studies prove a genetic link vs. the conditions of the womb being the cause? Until we create an artificial wombs all twins will share the same womb conditions, especially identical twins who often share the same amniotic sac as well.

17

Do environmental conditions only include the womb? No.

18

Have we?

19

[QUOTE=eschereal]
Wait, do I understand SNP to mean CG|AT|GC|TA ?
[/QUOTE]

Not sure what you mean by that. A SNP is a “spelling change” in a single base pair.

23andMe doesn’t read your genome, it does a key word search on it. With a limited set of key words. So while it may not be finding instances of “conformed bachelor” in your DNA, they not have even looked for “friend of Dorothy.”

20

John Mace answered you the first time. By comparing them to fraternal twins.