Nature versus Nurture? It’s a century-and-a-half-old debate that most sources credit to Sir Francis Galton. Today we know that behavior is determined by both our biological make-up and environment, yet many questions still remain. For instance, which biological and/or environment factors actually influence behavior? You might think this to be a fairly simple and easy to answer question, yet scientists don’t really have a clear answer. Nonetheless, studies such as the one recently published by Dr. Ganna and colleagues inevitably bring us closer to understanding the biology of human sexuality.
It was last October, at the American Society for Human Genetics Annual Meeting, where I first learned of Dr. Ganna’s work. His talk explored the results of a large-scale study on the genetic architecture of same-sex sexual behavior. The controversial topic sparked many questions from the speaker’s scientific colleagues. Luckily for us, Dr. Ganna’s work was recently published in Science. The study is particularly exciting because of its large sample size (477,522 individuals). In fact, previous reports of genetic variants associated with sexual orientation were based on much smaller sample sizes. If you’re wondering how the researchers were able to get so many study participants, the answer is both the UK Biobank and 23andMe. The UK Biobank study includes genetic data on ~500,000 UK residents aged 40 to 70 years old, while the 23andMe sample included mostly American 23andMe customers who consented to have their data used in research. While their dataset was indeed large, the primary survey question only asked whether a person had ever had sex with someone of the same-sex. You can imagine that this type of binary categorization (a yes or no answer) collapses the rich and multifaceted diversity among sexual minorities. I’m sure you’ll also agree that sexual orientation is far more complicated than just who we have sex with. In any case, the authors were fairly transparent about these limitations.
Genome Wide Association Studies (GWAS)
So now that we understand who was part of the study, let’s talk about Genome Wide Association Studies (GWAS). Simply put, a GWAS is an observational study that looks at variants in the human genome and determines whether a variant is associated with a trait. In most GWAS studies, the variants that are being looked at are what geneticists call single nucleotide polymorphisms (or SNPs). Now to understand what a SNP is, you’ll have to know that our genome is made up of a simple four letter code (A, C, G, and T). Let’s say we look at a very specific site in the human genome and find that most people have the letter A there, but some have the letter T. If we then find out that most of our T people are more trustworthy than our A people, then we can say that this SNP is associated with a trait (i.e. trustworthiness). In summary, the authors basically looked across the entire genome and tried to determine whether any variants were associated with their trait of interest (same-sex sexual behavior).
Same-Sex Sexual Behavior in Families
Using their rich dataset, the authors first assessed whether same-sex sexual behavior clustered in families. When looking at pairs of related individuals (full cousins or closer) the authors found that more closely related individuals were more likely to be concordant in their sexual behavior. What that basically means is that if one brother is gay, then there is a higher likelihood that the other brother is gay as well (in comparison to if the first brother were straight). The authors also came up with a broad-sense heritability estimate. This is just a fancy way of saying that they tried to tease apart how much of the variation in same-sex sexual behavior could be explained by genetic versus environmental factors. What they found was pretty similar to other published estimates (i.e. 32.4% of a genetic contribution).
Sex Differences
Now looking between men and women, the authors discovered something that wasn’t entirely surprising but still quite interesting. They found that the genetic architecture underlying same-sex sexual behavior was only partially shared between the sexes. In fact, their sex-specific analyses revealed male and female specific variants that associated with same-sex sexual behavior. While looking at the functional relevance of these sex-specific variants, the authors found something peculiar. For male specific variants, the biological pathways that may be affected included olfaction (i.e. our sense of smell) and male pattern balding. The latter makes more intuitive sense than the former, since male pattern balding is likely related to hormone regulation. With respect to olfaction, some studies have previously linked olfaction to reproductive function.
Genetic Correlations With Other Traits
Beyond their primary results, the authors also found correlations between same-sex sexual behavior and traits like loneliness, risky behaviors, and various mental illnesses. It’s important to keep in mind, though, that these relationships may exist for environmental as opposed to biological reasons. Same-sex sexual behavior may, for instance, increase one’s risk of developing a mental illness through the pathway of minority stress. Interestingly, some of these correlations were stronger in one sex versus the other.
Beyond a Yes or No Answer
To look at the next portion of the study results, let’s revisit how the authors determined whether a study participant had engaged in same-sex sexual behavior. You’ll remember from the beginning of my post that they used a binary classification system.
Binary classification
Question = have you ever had ever had sex with someone of the same sex?
Answers = yes or no
You’ll also remember that this was one of the study limitations. Therefore, to broaden it up a bit, the authors next used a proportion of same-sex sexual behaviors as their trait of interest.
Proportion
Question = with whom have you had sex?
Answers = other sex only, other sex mostly, other sex slightly, equal, same sex slightly, same sex mostly, or same sex only.
You’d expect that those variants that originally associated with same-sex sexual behavior would be more strongly associated in those reporting a higher frequency of same-sex sexual behaviors. Yet this wasn’t the case. These findings suggest that there is no single genetic continuum from opposite-sex to same-sex sexual behavior. Said another way, variants are associated with the presence of ANY same-sex sexual behaviors, as opposed to the PROPORTION of same-sex sexual behaviors.
Conclusions
In my scientific opinion, I’d say that this study mostly reminded us that the genetic architecture underlying sexual orientation is fairly complex. It’s more likely the case that sexuality is influenced by many genetic variants as opposed to just a few. Even then, genetic architecture may only partially explain why an individual might engage in same-sex sexual behaviors. All in all, this study further strengthens the reality that the infamous “single gay gene” is nothing more than a misconception.