Monday, 27 October 2008

One more part of the puzzle

This time it's been over-hyped in the headlines, yet it's still significant. Another small piece of just exactly the right shape, and exactly where we predicted it should be.

The reporting has been spotty, none of the newspaper articles giving a complete picture, so I'll try to marshal the facts together, and the significant comments on those facts.

From The Australian:
'Transexual gene' found by researchers
Well, not as such... it's more complex than that. We haven't found THE gene for Transsexuality, merely A gene that makes it more likely. There will be more, and possibly more significant ones. But I'm getting ahead of myself.
In the largest genetic study of its kind, 112 male-to-female transsexuals took part in a study involving several Melbourne research bodies and the University of California, Los Angeles.

Researchers measured the variation in the androgen receptor gene, which is involved in the functioning of the sex hormone testosterone.

DNA samples from the transsexuals were compared with the samples from 250 typical men.
"We think these genetic differences might reduce testosterone action and under-masculinise the brain during foetal development," Prince Henry's Institute researcher Lauren Hare said.
Which is in accordance with everything I've written on the subject in this blog. But I'll give a summary of that later, trying to build an incomplete but coherent picture of what we think happens, and how we think it happens too - though there's a lot of handwaving in that part. The "what" we're far more sure of, the "how"... not so much.
Lead researcher, Associate Professor Vincent Harley, head of molecular genetics at Prince Henry's Institute, said there had long been debate about the causes of transsexuality.

"There is a social stigma that transsexualism is simply a lifestyle choice. However, our findings support a biological basis of how gender identity develops," he said.
To say the least...this is a highly charged and politicised subject, with vested interests on both sides obscuring rather than clarifying the subject. Big Religion has insisted that it's about Morality with a capital M. Big Science has eventually come to the conclusion that there's a biological basis, and this has been known about for some time. I'll quote the Full Bench of the Australian Family Court once more on the issue, as far back as 2003:
At paragraph [270]: ‘But I am satisfied that the evidence now is inconsistent with the distinction formerly drawn between biological factors, meaning genitals, chromosomes and gonads, and merely "psychological factors", and on this basis distinguishing between cases of inter-sex (incongruities among biological factors) and transsexualism (incongruities between biology and psychology)’.

At paragraph [272]: ‘In my view the evidence demonstrates (at least on the balance of probabilities) that the characteristics of transsexuals are as much “biological” as those of people thought of as inter-sex’.
Note the "on the balance of probabilities". That's especially important in this context, as the evidence from this experiment alone is nowhere near enough for proof. Seen in context with many other pieces of evidence though, many even stronger, I think the case now is unimpeachable. We can now be confident enough to feel surprised if future experiments don't show the same pattern.
He said researchers were recruiting transsexual people for another study and hoped to double the sample size and examine other genes.
I'll go into why this piece is a relatively small one later, but suffice it to say that while statistically significant, it's relatively weak. We need more data, lots more data, in this particular area to be certain.
Julie Peters, a transgender person, said she knew from as young as three or four years old she did not fit into being a boy.

"I have always had the personality of a girl, I suppose is the way I perceive it and even from a very young age, three or four, I was really mad at people for making me a boy," she said.

"I personally think it (gender) is a combination of both (nature and nurture).

"You are born with a predisposition to have a certain personality and then depending on the culture you are brought up in your personal situation it becomes active in a particular way."
Remember that quote, I'll be getting to that aspect later. All will be revealed.
Here's a similar view, from the Sydney Morning Herald:
The findings, which are published in the journal Biological Psychiatry, were good news, said Sally Goldner, 43, who had an inner sense she was female from a young age, despite being a boy. "Such compelling evidence dispelling the total myth of gender identity issues being a choice is always welcome," she said.
Knowing from such an early age is typical, by the way. In fact, there's a "standard transsexual narrative" that the majority of transsexual people conform to (though there are numerous exceptions and variations on the theme). For example, my own life history conforms pretty well - but I knew at age 9 rather than 4 or 5, and then there's the whole "natural sex change" bit late in life that is totally out of left field. But overall, good agreement with theory, apart from the last peculiarity.

Now let's have a look at the detail...
The Australian and American team examined three sex hormone genes. They found male-to-female transsexuals tended to have a longer version of the androgen receptor gene, which could reduce testosterone action.

"It is possible that a decrease in testosterone levels in the brain during development might result in incomplete masculinisation of the brain in male-to-female transsexuals," Associate Professor Harley said.

But it was highly likely that other genetic factors were also involved in this form of transsexualism, he said.
Now would be the best time to explain what we think happens. There's some handwaving and "pay no attention to the man behind the curtain" here, but actually less than in comparable cases. The effects of hormones during puberty are so well documented as to be indisputable for example, but we still don't have a good idea "how" things happen, the mechanism, even though the "what" is well understood. The handwaving there is of exactly the same kind, but there's even more of it.

The story as we see it today... all opinions subject to change as more data comes in.
In the womb, the combination of genes and hormones leads to sexual differentiation of both the body, and the central nervous system - the CNS - (just don't ask how it happens, Ok? It happens)

Usually, 98%+ of the time, everything's consistent. Usually. But either the "wrong" hormones, or a genetic peculiarity, can lead to anomalies. These will often be partial rather than complete, as development in the womb proceeds in several stages, and timing is everything. In the case of the CNS, it's even more complex, as rather than full-blown development happening, it sets up "gradients", "propensities" that future development will follow. The same thing happens with genitalia and other structures to a lesser degree, these will develop in certain ways during puberty, triggered by hormonal floods which are themselves pre-determined by development in the womb. So some sexual differentiation is immediately obvious from birth (and indeed before), but other parts aren't.

When things go wrong (I won't go into the argument as to what is a "natural variation" and what is a "defect"), pretty much anything can happen. Usually you get patterns, syndromes where multiple things go awry. For example, with CAIS - Complete Androgen Insensitivity Syndrome - you get a sterile female with an incompletely formed female reproductive system despite having 46xy chromosomes, but in (virtually?) every case, a standard female neurology. Less Transsexuality than in 46xx female-bodied people generally.

On the other hand, people with Swyer syndrome and 46xy chromosomes, despite having in many cases a fully functioning female reproductive system (bar only the ovaries) have high rates of transsexuality and androgeneity. The anomalous development is not as complete in the neurology as it is with CAIS, but more so elsewhere.

But yes, you can have people with typically feminine neurology in every respect but one - such as having a body image that requires male genitalia - or people with neurology that match neither male nor female, but something of both, or just plain neither. And this may or may not involve other Intersex anomalies too, the odds are increased that they will, but not to the point of certainty.

"Typically feminine" neurology does not of itself lead to a female gender identity. That is mostly a label, even a social construct. Mostly. A child with a male anatomy brought up in an atmosphere where other children with male anatomies also have typically feminine neurology, so have the same emotional instincts and thought patterns, will identify as male too. It's only when they contact the outside world and find out that everyone else with the same instincts are labelled as "female", and expected to play with barbie dolls not toy cars that they'd get severe cognitive dissonance.

Except it's not quite that simple - because if the instincts that are feminised include body image, there would be severe discomfort with what's between the legs. It would be indefinably and inexplicably wrong in a very basic way.

The point is though, that transsexual children aren't brought up in this way. 98%+ of those whose instincts and emotional response matches theirs wear different clothes and are designated as the opposite sex to their own assignment. And so they develop a cross-gendered gender identity compared with the one they've been assigned. You can modify their behaviour by "aversion therapy", but not the underlying instincts. It's inevitable, and will get stronger with age and greater socialisation.
There are many ways that things can go awry, and in different ways, so there cannot be a single "transsexuality" gene.

I can make a prediction though, or rather, two predictions.
  1. That every single gene involved in decreased effects of testosterone in the womb will be over-represented in Male to Female transsexuals. Sometimes grossly so - as in this case - and sometimes requiring huge sample sizes and subtle statistical analysis to find the difference.
  2. And that every single gene that involves increased effects of testosterone will be over-represented in Female to Male transsexuals (as I've blogged about before)
Testing these predictions may not be easy though, so they're less useful than they appear. "Grossly" means "maybe 10% more, maybe double". Statistically significant from sample sizes where n<1000 and n>100, no more than that. From Reuters:
The scientists collected DNA samples from 112 male-to-female transsexuals and found that they were more likely to have a longer version of the AR gene than another group of 258 non-transsexual men.

The longer AR gene was found in 55.4 percent of people in the
transsexual group and 47.6 percent of the non-transsexual men, they
wrote in an article published in Biological Psychiatry.

Samples were also analysed for two other genes, but no significant
differences were found between both groups.
That's enough for it to be very unlikely that it's a coincidence, but not exactly eye-poking-outedly obvious. More data is needed.

Let's look at the effects:
The researchers said the longer AR variant gene may have resulted in less effective testosterone signalling, a mechanism that masculinises the brain during early development.

"It is possible that a decrease in testosterone levels in the brain
during development might result in incomplete masculinisation of the
brain in male-to-female transsexuals, resulting in a more feminized
brain and a female gender identity," they wrote.

One of the researchers, Lauren Hare, said: "We think that these
genetic differences might reduce testosterone action and under
masculinise the brain during foetal development."
Exactly. Finally, to summarise, from the BBC:
Terry Reed from the Gender Identity Research and Education Society said she was convinced of a biological basis to transsexualism.

"This study appears to reinforce earlier studies which have indicated that, in some trans people, there may be a genetic trigger to the development of an atypical gender identity.

"However, it may be just one of several routes and, although it seems extremely likely that a biological element will always be present in the aetiology of transsexualism, it's unlikely that developmental pathways will be the same in all individuals."
All of this is in accord with similar patterns seen in puberty. It accounts for the observed phenomena in Intersex, Transsexuality, and even (most controversially) Transgenderism. And yes, Heterosexuality, Homosexuality, Bisexuality and different degrees in all three too. I really want to avoid that particular can of worms, having enough trouble with political fanatics in the IS and TS communities, but it looks like I can't.

One disclaimer: I'm with John Maynard Keynes in one respect - I'll quote him.
When the facts change, I change my mind. What do you do, sir?


RadarGrrl said...

...and I thought it was going to be about the transsexual chicken.

Hayley said...

'Until' another gene is found to add to the affect of the AR gene then it's still inconclusive and amounts to nothing.

Unknown said...

Zoe, I have yet in any of the articles on this study seen an actual citation of the paper. I would like to try and read it for myself(I'm funny that way). I did see something about it appearing in Biological Psychiatry, but cannot find the paper in recent issues up through Nov 2008. Do you have a citation for it?

Anonymous said...

Wow, DES affects chromosomal structure?

Chris said...

Fascinating! I can only hope that more and more of this comes out in my lifetime; it feels like the 'answer' gets closer all the time.

Zoe Brain said...

anonymous - Re DES - it's like Thalidomide, where both a pre-disposing gene and also environmental factors (ie administration of the drug) are necessary for developmental anomalies to occur.

So no, chromosomes don't get altered, but the genetic makeup of those affected by Thalidomide and those not affected are different.

Zoe Brain said...

Shannon -

Androgen Receptor (AR) Repeat Length Polymorphism Associated with Male-to-female Transsexualism. Authors: Lauren Hare, Pascal Bernard, Francisco J. Sanchez, Paul N. Baird, Eric Vilain, Trudy Kennedy and Vincent R. Harley. The research will published in the Jan 2009 edition of Biological Psychiatry. Advance online publication date 27 Oct 2008.

I'll get you a URL or copy of the abstract when I get back to my laptop.

Zoe Brain said...

I did find a copy of the press release:

Under strict embargo until 00.01am Monday 27 Oct (AEDT)
Genetic link to gender identity
In the largest ever genetic study of male to female transsexuals
Australian researchers have found a significant genetic link between
gender identity and a gene involved in testosterone action.

From an early age people develop an inner sense of being male or
female – their gender identity. Transsexuals however, identify with a
physical sex opposite to their perceived biological sex.

DNA samples were collected from 112 male to female transsexuals and
researchers compared genetic differences with non transsexuals. The
results are published in the high impact journal Biological

The researchers discovered that male to female transsexuals were more
likely to have a longer version of a gene which is known to modify the
action of the sex hormone testosterone.

"We think that these genetic differences might reduce testosterone
action and under masculinise the brain during foetal development."
said researcher Lauren Hare.

For decades, there has been debate over the causes of transexuality.
Early theories included psychosocial factors such as childhood trauma.
More recent studies have indicated that family history and genetic
aspects are linked to the development of gender identity.

"There is a social stigma that transsexualism is simply a lifestyle
choice, however our findings support a biological basis of how gender
identity develops." said study leader Associate Professor Vincent
Harley, Head of Molecular Genetics at Prince Henry's Institute.

"As with all genetic association studies it will be important to
replicate these findings in other populations" said Associate
Professor Vincent Harley.

Researchers are now planning even larger genetic studies and are
investigating a wider range of genes that may be related to gender

Online briefing for journalists
Friday 24 October 10.30am AEDT
Contact for further information

Further Information

Androgen Receptor (AR) Repeat Length Polymorphism Associated with Male-
to-female Transsexualism. Authors: Lauren Hare, Pascal Bernard,
Francisco J. Sanchez, Paul N. Baird, Eric Vilain, Trudy Kennedy and
Vincent R. Harley. The research will published in the Jan 2009 edition
of Biological Psychiatry. Advance online publication date 27 Oct 2008.

This international study involved researchers at Prince Henry's
Institute, Monash Gender Dysphoria Unit, Monash University, Melbourne
University, the University of California, Los Angeles in the USA.

The research was funded by the National Health and Medical Research
Council Australia and the US National Institutes of Health.

For more information contact Ian Muchamore, Science Communication,

About Prince Henry's Institute

Prince Henry's Institute (PHI) is committed to improving the quality
of life through the investigation of hormones and has a team of 130
people dedicated to scientific excellence.

PHI is world renowned for its research into reproduction and
endocrinology, the study of hormones. The Institute's research focuses
on the role of hormones in cancer, reproductive health, diabetes,
obesity and heart disease .

Established in 1960 as the Medical Research Centre at Prince Henry's
Hospital in South Melbourne, PHI is now an independent Institute based
at Monash Medical Centre in Clayton, Melbourne, Australia.

It is funded by federal and state government grants and supported by
individual, business and community groups


Zoe Brain said...

Abstract here.

baglieg said...

Hi Zoe!

The gene in question has 20-something different alleles. Trans people don't all have the same allele, so the sample size was too small to make any statistically significant findings with any one allele.

So the researchers arbitrarily divided them into two groups according to how many base pairs each allele has, allowing them to draw some limited conclusions from such a small sample.

55.4 percent of transsexuals and 47.6 percent of the non-transsexuals had alleles from the group of longer alleles.

However, since the vast majority of people are not trans, that means that most people who have one of the longer alleles are not trans.

I did a rough back-of-the-envelope calculation, and found that if 1 in 500 people are trans, then that would mean that having an allele from the longer group would increase the chances of an XY individual being trans by approximately 0.06%. Or even less if there are fewer trans people in the population (at an absolute minimum, by 0.001%).

It would take a larger study to look at the effects of individual alleles to find out whether there is one allele that produces, say, a 1% increase in the chance of growing up trans, or whether this miniscule effect is just a blip in the numbers that was inadvertently magnified by selectively grouping the alleles in the way which produced the greatest difference.


Zoe Brain said...

In other words... more data needed. We have a 96% chance that there's something, enough to fund further research, but that's all.


Thanks Robbie.

Unknown said...

Thank you. That explains why I couldn't find the paper. Science Direct only has up through Nov. I will wait for the Jan to come out.

Nichole said...

There are times, Zoe, that I would like to have your scientifically-inclined brain in my own head. But, then I find I am perfectly satisfied with my own and so send people to you for the science. :)

I found your blog perfectly readable for one trained in Lit & not in scientific writing.

You do and have done a fantastic job getting out the new studies and encapsulating them so even an English honors graduate can follow them.

Keep up the excellent posts and writing. You matter so very much for all of us, cis and trans.

Thank you.


Anonymous said...

Maddie H said...

Yeah, we can always rely on freepers to be a bastion of ignorance and intolerance.

They're fucking clueless.

Anonymous said...

So, after years of telling my mother "it's nothing to do with you or dad" now I have to tell her it was genetic?

Since the genes came from them she's definitely going to interpret this as "it's all our fault"!


Unknown said...

While I do think this takes another baby step to establish that biology at least plays a partial role in causing transsexuality, it is still very unclear to me if nailing down a cause is possible.

There needs to be a lot more research done to understand environmental impacts on genetic code. Consider how little we understand cancer and its relationship it to genetics. Let alone a behavior as complicated as transsexuality, which is surely on the order of a dozen times more complex in terms of symptoms.

Anonymous said...

Regarding the paper in Biological Psychiatry, you can see a discussion, citation, and abstract here:

Genomic Association Study Identifies Genetic Basis for MtF Feminization