We're Probably Not Getting Dumber

There's been buzz over the past few days in the idea that humans have been getting stupider for the past few millennia. That's according to Stanford's Gerald R. Crabtree -

I would wager that if an average citizen from Athens of 1000 BC were to appear suddenly among us, he or she would be among the brightest and most intellectually alive of our colleagues and companions, with a good memory, a broad range of ideas, and a clear-sighted view of important issues. Furthermore, I would guess that he or she would be among the most emotionally stable of our friends and colleagues.

OK. How so?

Crabtree's argument in a nutshell:

In Part I, he outlines the latest evidence showing that many thousands of genes contribute to human cognitive ability, and that because mutation rates are high (higher than previously believed), any given individual probably carries harmful variants of many of these genes. This is quite possibly true ,and interesting, but by itself it's nothing to do with declining IQ.

In Part II, Crabtree says that during human evolution, all of these genes were under strong selection pressure because any human or proto-human who wasn't smart enough to hunt, fight and survive in the stone age environment, would get eaten by a predator or starve. However, after these hunter-gatherer tribes became settled farming communities (in say 6000 BC), they were no longer so vulnerable, so the less intellectually able could live... and breed... leading to ever-more unintelligence genes.

Now, there's a lot of problems here. Many have said that this is all a bit like eugenics, and indeed it is, but that doesn't necessary mean it's wrong... no, it's wrong because the argument is flawed.

For instance, it's already been pointed out that, even if it recently got easier to stay alive, that doesn't mean it's got easier to get laid lots and have lots of kids; and sexual selection is a powerful force in evolution, perhaps even stronger than survival, and it probably favours higher intelligence.

However, there are other issues.

The idea that hunter-gatherers have especially hard lives is dubious. There's good evidence that life expectancy and health fell when hunter-gatherer societies settled down and got agriculture. Today, survival doesn't exert much of a selective pressure in most parts of the world but life stayed pretty precarious (by modern standards) until at least the 19th century.  You could indeed argue that 8,000-odd years of agriculture made us smarter than ever before, and that we're enjoying the benefits . In which case we'd be smarter than someone from 1000 BC, who only had 5,000 years or so.

Crabtree does acknowledge that agriculture changed selection pressures. He suggests that it would have made it more important to be immune to the various diseases that emerged when our population density rose. But this isn't enough - for his argument to work, intelligence would also have needed to become less important with agriculture, and whether that's true is really not clear at all. Maybe it did. Maybe it didn't. There's a prejudice in modern culture against 'ignorant peasants' and 'dumb hicks', but farming is not easy.

Even if we do grant that cognitive evolutionary pressures have eased since 1000 BC, it's not clear that this would make us 'less intelligent'. 'Intelligence' is not one thing. To simplify, it might be that there's a trade-off between 'book smarts' and 'street smarts', and that you used to need the latter to survive. A society in which everyone survives would then allow more people the luxury of being book-smart. Would Einstein or Newton would have made good peasants?

Crabtree's arguments are interesting, but they're entirely speculative. There's just no hard evidence for the decline of intelligence over recent millennia. Since we can't go back in time and do IQ tests, there never will be, although he does suggest an experiment, using genetics, that might be able to check whether there's been a build-up of harmful mutations.

But until then, as he puts it,

in the meantime I’m going to have another beer and watch my favorite rerun of ‘Miami CSI’ (if I can figure out how to work the remote control).

Crabtree, G. (2012). Our fragile intellect. Part II Trends in Genetics DOI: 10.1016/j.tig.2012.10.003 

Crabtree, G. (2012). Our fragile intellect. Part I Trends in Genetics DOI: 10.1016/j.tig.2012.10.002

Gene-Guided Antidepressants?

Over the past couple of years, "Big Pharma" has largely moved away from psychiatric drug development. This shift has been widely discussed.

But another trend has been happening over the same time period - or so it seems to me. This is the rise of small companies who offer techniques for diagnosing mental illness, or predicting which drugs will work best. Generally (it seems) partnerships between venture capitalists and psychiatry (ex-)researchers, these enterprises might be dubbed "Little Pharma".

The latest is a company called AssureRx Health, Inc. According to a paper just published, they offer

a pharmacogenomic algorithm designed to improve the safety and efficacy of prescribing antidepressant and antipsychotic medication... based on the genotyping of both copies of five genes.

From this, you end up with a report giving each drug a rating of green, yellow, or red (see above).

The price is not provided on their website.

According to the paper, they gave 26 depressed patients normal treatment at the discretion of their psychiatrist, while 25 got treatment guided by the AssureRx algorithm. It was non-randomized, and unblinded so there's a clear possibility of a placebo effect.

Anyway, the results were...

For the first 4 weeks of treatment, there was no difference between the two groups whatsoever in terms of depression symptom scores - they both improved. But then by week 8, the unguided patients abruptly got worse, while the AssureRx-guided ones continued to benefit. This is an unusual pattern of improvement in an antidepressant trial.

Previously, I wrote about another Little Pharma antidepressant prediction scheme. It used a  different approach, measuring brain electrical activity using a technique called "rEEG", rather than genetics. But the basic idea is the same... and so are the problems.

As I said last time:

There were two groups and they got entirely different sets of drugs. One group also got rEEG-based treatment personalization. That group did better, but that might have nothing to do with the rEEG: they might have done equally well if they'd just been assigned to [those drugs] by flipping a coin. We cannot tell, from these data, whether rEEG offered any benefits at all.

In the AssureRx paper, we can't even tell whether the two groups got different kinds of drugs, because the meds used aren't reported, but if they did differ then that would offer an alternative explanation for the differences in outcome: maybe the 'guidance' just recommended better drugs overall, with the genes being just a sideshow. Or maybe it's a placebo, as I said.

Moving on, I also wrote... 

What's curious is that it would have been very simple to avoid this issue. Just give everyone rEEG, but shuffle the assignments in the control group, so that everyone was guided by someone else's EEG. So you'd give control Patient 2 the drugs that Patient 1 should have got, and vice versa; swap 3 and 4, 5 and 6, etc.

This would be a genuinely controlled test of the personalization, because both groups would get the same kinds of drugs... and it would allow the trial to be double-blind: in this study the investigators knew which group people were in, because it was obvious from the drug choice...

It is odd that Little Pharma never seem to do such real vs. muddled prediction studies, as they'd be really informative as to whether their approach is a helpful innovation as opposed to an expensive, meaningless red herring. Hmm.

Hall-Flavin, D., Winner, J., Allen, J., Jordan, J., Nesheim, R., Snyder, K., Drews, M., Eisterhold, L., Biernacka, J., and Mrazek, D. (2012). Using a pharmacogenomic algorithm to guide the treatment of depression Translational Psychiatry, 2 (10) DOI: 10.1038/tp.2012.99

Brains In A Dish Need Sleep Too?

All animals sleep, but despite decades of research, neuroscientists still have no clear answer as to why. Now a dramatic new study reveals that sleep may be a fundamental state that even brain cells growing in a dish need.

Swiss neuroscientists Valerie Hinard and colleagues cultured mouse cortical neurons in dishes equipped with arrays of electrodes. This allowed them to record the electrical activity produced by the growing 'brain'. They also measured the expression of different genes in the neurons, and compared these to gene expression in real mouse brains.

They found that while cultures of neurons started out fired randomly, after about 10 days, the cultures entered a state of synchronized periodic firing, with the whole population of cells firing together in slow cycles of activity - with a frequency of 1 cycle every 5 to 15 seconds. This is extremely slow - by contrast the "slow waves" characteristic of animal sleep cycle about 30 times faster - but the authors say that such ultra-slow waves have been seen in sleeping animals too.

But the dishes could be 'woken up' by adding a mixture of neurotransmitters, which abolished the burst cycles. They reappeared about 24 hours later. Gene expression changes in the cells in the 'sleep' and 'wake' state were significantly correlated with changes seen in real mice deprived of sleep.

Finally - and this might end up being the most important bit - the authors compared the biochemistry of the 'sleep deprived' dishes to the 'well rested' ones. They found remarkably few major changes, but they did observe a significant increase in the levels of lysolipids.

Lysolipids are breakdown products of phospholipids, which make up the membranes of all living cells. When present in membranes, lysolipids can act as 'detergents', distorting their structure. That's bad. These results suggest that sleep might serve to prevent the build up of lysolipids. If that pans out, it would mean that the function of sleep is very primitive, a fundamental biological necessity for any connected network of neurons, even what amounts to a random medley thrown together on a plate.

This study used cultured mouse neurons, but it's possible to grow human brain cells in a dish too. The obvious next step will be to check if human neurons exhibit the same sleep/wake-like states - and whether the very slow synchronized firing is really like human sleep. If so, could this help understand insomnia? Narcolepsy? Maybe even jetlag?

It's also got implications for all other brain-in-a-dish research. Scientists may literally need to ensure that their dishes get enough sleep in future studies.

It's all very exciting. I don't study sleep in my own research, but I try to keep up with the literature as I find it very interesting. I've covered various aspects of sleep neuroscience previously. So while I'm no expert, this seems to me like truly groundbreaking stuff, and potentially a "game changer" for the whole of neuroscience.

Hinard V, Mikhail C, Pradervand S, Curie T, Houtkooper RH, Auwerx J, Franken P, and Tafti M (2012). Key electrophysiological, molecular, and metabolic signatures of sleep and wakefulness revealed in primary cortical cultures. The Journal of neuroscience : the official journal of the Society for Neuroscience, 32 (36), 12506-17 PMID: 22956841

A Case Study in Voodoo Genetics

A new review of published studies looking at the relationship between a gene and brain structure offers a sobering lesson in how science goes wrong.

Dutch neuroscientists Marc Molendijk and colleagues took all of the studies that compared a particular variant, BDNF val66met, and the volume of the human hippocampus. It's a long story, but there are various biological reasons that these two things might be correlated.

It turns out that the first published reports found large genetic effects, but that ever since then, the size of the effects has dropped, with the latest studies finding no effects at all -

A cumulative meta-analysis confirms that as more studies on BDNF val66met have appeared, the overall effect estimate has steadily declined -

Finally, the authors found signs of publication bias: there were three small, imprecise studies that reported very large effects of the gene, but no such studies finding no effect (or a reverse effect). You'd expect that small and noisy studies would have a lot of random variation so they wouldn't all be positive even if there was a true effect; that all of the published ones were positive, suggests that null findings are out there, unreported.

Overall, this suggests that val66met probably isn't associated with hippocampus volume after all, and that the early studies showing that it was, were misleading. There's no reason to think that the early studies were wrong as such - they may have accurately reported an effect in the small sample of people they looked at, but it was only a chance finding.

I know a lot of neuroscientists who are now fairly skeptical of this whole genre of candidate gene studies; there was much excitement 5 or 10 years ago, but in retrospect, most of these studies were too small, and the publication process meant that it was the random chance findings that were most likely to get published.

However, we need to avoid any sense of complacency. Until we fix the scientific process, the same thing will happen again.

Molendijk ML, Bus BA, Spinhoven P, Kaimatzoglou A, Voshaar RC, Penninx BW, van Ijzendoorn MH, and Elzinga BM (2012). A systematic review and meta-analysis on the association between BDNF val(66) met and hippocampal volume American journal of medical genetics B Neuropsychiatric genetics PMID: 22815222

The Coming Age of Fetal Genomics

It's 2020. A young woman and her partner have just found out that she's pregnant with her first child. and they're going to be parents.

They're overjoyed, of course. But they're also worried. They've seen the adverts warning parents-to-be about the risk of de novo mutations - genetic mistakes that occur inside sperm or egg cells, and affect the child. These mutations, the ads say, are much more common than previously believed and they can cause all kinds of problems: intellectual disabilities, autism, infertility, mental illness, heart malformations - pretty much anything.

The scariest part? Because these are new mutations, out of the blue, they can affect any family. A clear family history is no protection. They don't discriminate by race or lifestyle. It's just the luck of the draw - except that older parents are at much higher risk, especially older fathers. In the case of our couple, she's 28 and he's 32. Perfectly normal for this day and age - but very old in biological terms. Humans evolved to be grandparents by 32, not parents. "The stakes couldn't be higher. Why leave it to chance?"

So they don't. Instead, they buy a $100 test kit, they each provide a small blood sample and send it off to one of the companies offering fetal genome testing. At the testing lab, they can separate out the mother's DNA from that of the fetus, both of which are present in the mother's blood. By comparing the fetal genome to the mother's and father's, it's easy to spot de novo mutations. If a certain gene doesn't match either the mother or the father's sequence, it's mutated.

A few days later the results are back. There are several mismatches detected. Most are benign - they're not predicted to have any biological effects. But there's one, a deletion of a few thousand bases in a gene involved in brain development. This deletion is predicted to raise the risk of epilepsy and autism from 1% to 10% apiece.

The parents now have a decision to make. The mutation is a one off, it's not inherited. If they conceive again... roll the dice again... and it'll be gone. Do they terminate?

Like the adverts say, "Some people disagree with this, but we say there's only one person who really matters: your baby."

*

This is likely to become possible in the next few years.

A paper just published in Nature reports on the Non-invasive prenatal measurement of the fetal genome. The technique relies on the fact that the blood of a pregnant woman contains DNA: hers, obviously, but also that of the unborn child. This cell-free DNA can be extracted and genotyped.

This has been possible for a few years, but until now, only fairly crude genetic information could be detected. An extra chromosome, such as in Down syndrome, is pretty easy to spot. This technique is already used to diagnose Down's syndrome and a few other disorders prenatally.

But those diseases are just the low-hanging fruit at the tip of the iceberg, if you see what I mean. To gather the kind of detailed genomic information that could diagnose thousands of disorders is harder: the fetal and maternal genomes are mixed up, and the challenge is to tease them apart. But according to the Stanford geneticists behind the Nature paper, and other teams, it's now possible as early as the first trimester.

What makes the new method so revolutionary is that it is, as the title of the paper says, non-invasive. It's already possible to sequence a fetal genome, but it takes a surgical procedure involving inserting a needle into the womb, and a degree of risk. It's not something you can just sit down and do - but blood samples are. So cell-free DNA will make fetal genomics a personal choice, a commercial product.

The resolution's still not 100%, but inevitably, it will become cheaper, faster and more accurate as technology advances. This year we're expected to see the cost of reading a whole human genome falling below $1000. Fetal genomes will be more expensive, but not enormously so. In 5 or 10 years, it's likely to be affordable.

*

What will happen? I think there'll be demand for such services. Most parents won't do it, but enough people will that it will be a major issue. Just look at countries where boys are more valued than girls: a lot of sex-selective abortion happens. Today, it's limited to gender, because it's easier to determine a fetus's sex than its genome. In 5 or 10 years, they'll both be easy.

With demand will come companies to supply these services and, inevitably, advertizing. I doubt these adverts will be on TV, because there will be opposition to the whole idea and boycotts of broadcasters who run them. But we'll be getting spam emails about it. People will worry even over the harmless mutations: a variant won't need to be really associated with a disease, just believed to be, to make people panic.

Socially, it's likely to be divisive. The anti-abortion types will obviously not approve. But it will drive a wedge between those who support abortion but oppose "discriminatory" abortion against the disabled, and those who support the right to terminations 'for any reason, or none'.

Politically, there will be pressure to regulate this, but legally, it'll be tricky. I can't see how you'd prove that any given abortion is motivated by genetic concerns, so unless you ban abortion outright, that won't work. Banning fetal (or all) genomics except under medical supervision might be possible, but people could always go (or send a few drops of blood) abroad to get around that.

I'm not sure where I'll stand on this, but it looks very likely that we'll each have to make a decision in the coming years.

Fan HC, Gu W, Wang J, Blumenfeld YJ, El-Sayed YY, and Quake SR (2012). Non-invasive prenatal measurement of the fetal genome. Nature PMID: 22763444

Identical Twins, Different Lives

Virginia psychiatrists Kendler and Halberstadt describe a neat "natural experiment" into what causes depression - The road not taken: life experiences in monozygotic twin pairs discordant for major depression

They interviewed 14 pairs of identical twins. One of each pair had reported a history of depression while the other hadn't. The twins were interviewed together, and asked to describe their lives, in particular any differences between their experiences.

It's well worth reading, for the human interest stories if nothing else. Here's perhaps the most striking one:

Lisa (never depressed) and Leslie (depressed), interviewed at age 53, were identical twins... they were together constantly as children, but described their personalities as somewhat different from the start... Lisa described herself as getting quickly upset over adversity, but then rapidly changing gears and focusing on problem-solving. Leslie indicated that she had more of a temper and was more assertive, more likely to ‘mouth off’ and get into trouble than her twin.

Lisa knew she wanted to be a school teacher, attended a teacher’s college and has taught for her entire career. Leslie was less certain of her career goals and held a number of different jobs... Lisa, at 24, met her husband of nearly 30 years and had a big, traditional wedding. Leslie married the guy she dated in high school and college, invited only their parents to the wedding and divorced after 5 years, commenting "I picked the wrong man."

Late in the interview, Leslie reported (only after being asked about the most difficult time in her life) that just over 30 years ago, after drinking a modest amount of alcohol, she became pathologically intoxicated and drove onto a major highway off ramp going the wrong way. She got into the highway going against traffic and had a head-on collision, which killed the other driver - a woman with young children. Leslie was not seriously injured.

Recounting this event was clearly difficult for her as she openly wept in telling us this story even after all these years. In recounting her psychological reaction to the accident, she said, "Am I really to blame for this, and then I’ll have to live with this for the rest of my life? knowing there was somebody else whose family had been destroyed there." The depression following this episode was her most severe and she talked about her deep sense of guilt. Although manslaughter charges against her were dropped, there was evidence that Leslie was at fault.

The other stories were less extreme, but no less dramatic in their own ways. The authors conclude that, of the 14 pairs, the depressed twin got depressed because of: romantic difficulties (7); single traumatic events (2, including Lisa & Leslie); employment difficulties (1), a mixture of factors (2), and for no clear reason at all (2).

However, what we can really conclude from all this? The study specifically took identical twins, who grew up together but only one of whom reported depression, so it ruled out genetic influences and also psychological and social factors shared by both twins - things like family background, growing up in poverty, etc. That's the whole point of the study but it's important to remember that these are unusual cases.

Beyond that, it's hard to know if the differences in the twins' lives caused the depression. All we know is that they're correlated. "Leslie" for instance became seriously depressed after causing the death of a woman in a drunk driving accident. A straightforward case of cause and effect, perhaps, but then why did she drink and drive in the first place? Was depression, or something associated with it, part of the reason?

Ideally, I would want to repeat this study in identical twins both (or neither) of whom had depression to see if their lives - before the depression at any rate - were more alike than these discordant twins. However, it's still a fascinating study.

Kendler, K., and Halberstadt, L. (2012). The road not taken: life experiences in monozygotic twin pairs discordant for major depression Molecular Psychiatry DOI: 10.1038/mp.2012.55

Gaydar Works (A Bit, On Facebook)

The media are gleefully reporting a recent paper showing that "gaydar is real" - we can tell who's gay just by looking: The Roles of Featural and Configural Face Processing in Snap Judgments of Sexual Orientation

While it's a fine paper, I'm afraid that the results really aren't that exciting.

American undergraduate students were able to classify people as gay or straight with better than chance accuracy, based purely on photos of their face. For male photos, the hit rate was 0.57; for women it was better with an accuracy of 0.65.

However, that's on a scale where you get 0.50 by flipping a coin. So saying that gaydar is '65% accurate', as almost everyone has, is misleading. Still, the numbers seem solid. The sample sizes were large and the effect was replicated very convincingly in two experiments.

However... this tells us very little about real world "gaydar", and it wasn't intended to. There are reasons to think it could underestimate the accuracy:

• Most importantly - people only saw the pictures for 50 milliseconds each. 1/20th of a second. Followed by a backward mask. That's right on the threshold of conscious perception, almost 'subliminal' but not quite. With longer viewing times, they might have done better.
• All the faces were black and white photos with the hair and ears cropped out (see above - and I think those two photos from the paper are the authors, although I may be wrong!). Anyone with facial hair, glasses, or any other 'accessories' wasn't used. In the real world, we have that extra information.
• In real life, we get clues from facial expressions, body language, voice, clothes. You could argue that these are being used (consciously or not) specifically as signals of sexuality, so they don't count as 'gaydar' - but more on that later.

 But it could also overestimate gaydar's powers:

• These were photos that people chose for their Facebook profiles. We all know how much effort some people put into that choice. We also know that different photos of the same person can often seem like two different people. Your Facebook pic is probably the most "selected" photo of you in existence. It would be better - but also much harder - to use passport photos.
• All of the gays in the study were out of the closet: they broadcast their sexuality on Facebook. But lots of gay people don't do that. Now those cases are probably where 'gaydar' is most likely to be of interest to most people, I think; those people might be harder to spot.

As far as I can tell, this study wasn't intended to "prove that gaydar works". It was meant to examine how it works, by seeing whether it works very quickly (yes - in 50 ms in some cases). The authors also tested how accuracy was changed by flipping the photos upside down; this reduced accuracy but it was still well above chance.

Ultimately, we need to ask what "gaydar" means and why we find it so interesting.

On a superficial level, it just means being able to sense, from someone's appearance, if they're gay. That certainly does 'work' - if you see a guy coming out of a gay club in a tight pink Boy George t-shirt then yeah, he's probably gay. But he's (effectively) told you so, by being in that club and wearing those clothes, so that's not very interesting. That's an extreme case, but clearly people advertise their sexuality (and much else of course) all the time. Gaydar, in a weak sense, is just perception.

I think what makes "gaydar" intriguing is the stronger idea that it can go beneath such adverts. That we can see who's really gay, whether or not they admit it, even to themselves. If that were possible, then it would seem to mean that homosexuality is part of the essence of some people - in other words, that it's a biological trait.

So gaydar in a strong sense is risque. It calls to mind un-PC ideas such as physiognomy and would seem to validate various stereotypes which are the stuff of dirty jokes more than polite discussion.

Does gaydar in this strong, exciting sense exist? That's another question. This study doesn't tell us.

Tabak, J., and Zayas, V. (2012). The Roles of Featural and Configural Face Processing in Snap Judgments of Sexual Orientation PLoS ONE, 7 (5) DOI: 10.1371/journal.pone.0036671

The Geography of Faces

How much can you tell about where someone comes from, just from their face?

The other day I was in London and came across a group of young people in Muslim attire who were waving (or in some cases wearing) a particular flag. I thought it was the Iranian flag, but, I thought, they didn't look Iranian. They looked more like Somalis, but it certainly wasn't the blue and white Somali flag. I decided that maybe they were some kind of pro-Iranian demonstrators, but I later worked out that it was the flag of the unrecognised state of Somaliland.

This got me thinking about how reliable these "they look they're from..." judgements are.

Clearly on a basic level, we can usually tell which continent someone's ancestors were from, in terms of the familiar "races" of Europeans, Africans, East Asians etc. But what about shorter distances?


Could you tell, just from looking at them (and setting aside dress, hairstyle, jewellery etc.) whether someone was from Spain as opposed to France? Korea or Japan? Russia or Germany?

I can only speak for England, but there's certainly a vague but widespread belief that every part of Europe has a  distinct 'look'. In the past, people were very fond of talking about that kind of thing; today, we're rather embarrassed by the idea but the belief lives on.

I don't know, but I'd be very surprised if there weren't analogous beliefs in other countries.

But how accurate are these folk beliefs, really?

Supposing you were the world expert on human faces - or suppose you were a supercomputer with face-recognition software and access to Facebook's entire dataset. How accurately could you place someone's origins on the map, on average? To within 1000 km? 100? With what degree of accuracy? In an ideal world, could the ultimate face-placer judge someone as French vs German 75% of the time? 90%? Or only slightly better than chance?

I suspect that if you researched this, you'd find that a supercomputer could do very well, in most parts of the world, but that the majority of actual people are less accurate than they think they are.

Personality Without Genes?

According to a paper just published (but available online since 2010), we haven't found any genes for personality.

The study was a big meta-analysis of a total of 20,000 people of European descent. In a nutshell, they found no single nucleotide polymorphisms (SNPs) associated with any of the "Big 5" personality traits of Neuroticism, Extraversion, Openness to Experience, Agreeableness and Conscientiousness. There were a couple of very tenuous hits, but they didn't replicate.

Obviously, this is bad news for people interested in the genetics of personality. But I wonder if the implications are even wider -

We know that there are SNPs associated with physical traits like height, weight, hair colour, eye colour, and the risk of various diseases. If none of those SNPs are associated with personality, then none of those traits are causally associated with personality.

"Short man syndrome"? A myth. Rod Stewart was wrong about blondes. There's no such thing as a "fat personality". And so on. Maybe that's not surprising, but more generally, the implication would be that the genes we inherit have no direct or even indirect influence on our personality, which is a pretty radical conclusion when you think through it.

I'm making some assumptions here. Maybe some genes are correlated with personality, but the currently popular "Big 5" approach is just a poor way of measuring of personality. It could also be that there are so many interacting genetic and environmental effects on personality that any given effect is tiny by itself, and even bigger sample sizes, or multivariate data analysis, would be needed to detect such effects.

de Moor, M., et al. (2010). Meta-analysis of genome-wide association studies for personality Molecular Psychiatry, 17 (3), 337-349 DOI: 10.1038/mp.2010.128

Genes for Intelligence - Back to Square One

Here's a paper - soon to appear in Psychological Science - which says that Most Reported Genetic Associations with General Intelligence Are Probably False Positives

The authors tried to replicate published associations between particular genetic variants (SNPs) and IQ (specifically the g factor). They looked at three datasets, a total of about 10,000 people, and didn't confirm any of the 12 associations.

As Razib Khan says in his post on this, "My hunch is that these results will be unsatisfying to many people." I'd go further and say that no-one will be happy with these.

For those who believe that IQ is purely environmental and not genetic, any satisfaction they might feel will be short lived because these authors did replicate the recent finding that genetic variants explain about 50% of the variance in IQ. Looking at all SNPs together, there was a strong correlation between "genetic similarity" and similarity in IQ. That independently confirms what the much-criticized twin studies of IQ said - IQ is about 50% heritable.

But for people who do believe in the genetics of intelligence, this shows us that we have no idea what the genes are, and that everything published so far has been pretty much for naught.

There's another implication. We actually do know of many "IQ genes" in that we know genes that, when mutated, cause mental retardation (very low IQ).

Now many researchers have hoped that if a certain gene causes you to have an IQ of, say, 50 when it's completely deleted by a mutation, then more subtle variants in that gene would have minor effects on IQ. Maybe a variant that reduces expression of the gene by 10% would knock off 5 IQ points.

In other words, if big mutations cause big phenotypes, then small mutations in the same place ought to cause small phenotypes. It seems to make sense - but today's IQ literature shows that it's just not true.

That's not just a problem for IQ though. Take autism or ADHD, we know that there are rare, severe mutations that cause these conditions. Many people are hoping that common variation in the same genes might also be interesting - but if IQ is anything to go by, it won't be.

Perhaps this is not so surprising. Breaking your neck and becoming paraplegic is going to seriously impair your ability to play baseball. That doesn't mean that normal variation in baseballing skill has much to do with minor neck injuries.

Chabris, C. F. et al (2011). Most Reported Genetic Associations with General Intelligence Are Probably False Positives Psychological Science

Scientific Databases - or Filters?

A new online database called AutismKB offers a quick way to find the evidence linking genes to autism.

You can read up on it in a paper describing the project.

You can browse by chromosome or gene name, it includes data on all kinds of genetic variants from SNPs to CNVs and it gives each variant a score according to the strength of the evidence. I haven't had a chance to really tell how useful these scores are, but there's an option to create your own score based on how much weight you give different kinds of evidence. The dataset is huge although it doesn't seem to have been updated for a few months.

Overall, it's a new tool and there's sure to be bugs to iron out, but it seems like it could be very useful. I do worry though that this kind of database encourages misleading ways of thinking about autism genetics.

There are numerous genetic variants which have been strongly linked to autism, although none of them account for more a small proportion of cases because these variants are rare. But many (most, actually, is my impression) of them have also been observed in people with other symptoms ranging from ADHD to epilepsy to schizophrenia.

So searching a database of "autism genes" could encourage you to think that these were only autism genes, which is far from true. Genetics, it is becoming increasingly clear, doesn't respect our current concepts of psychiatric illness or our academic specialities. There are few (if any) parts of the genome that can be neatly fenced off and declared exclusive to ADHD experts, schizophrenia researchers or whatever.

But disease-specific databases encourage the illusion that they do exist. It's the same old problem of the filter bubble which many people have warned about in the context of general purpose search engines. Scientists have filter bubbles too.

This is not of course a criticism of AutismKB in particular - the same goes for any similar "disease-gene" database. And to be fair AutismKB does provide links to a schizophrenia database, and a couple of others but you have to dig quite deep to get there. The "main page" of results for any given variant is pure autism.

That's the whole problem with filter bubbles - they make it too easy to hear what you want to hear, compared to getting a new perspective, so you don't even think to look outside the filter.


Xu LM, Li JR, Huang Y, Zhao M, Tang X, and Wei L (2011). AutismKB: an evidence-based knowledgebase of autism genetics. Nucleic acids research PMID: 22139918

The Gene That's "For" Nothing

Scientists like to warn you not to talk about "the gene for" a particular disease or trait.

I've done so in previous posts e.g. this one or this one.

But such scalding is not always very effective. We like simple explanations, so we like to find simple connections between genes and phenotypes.

Which is why a new paper is important. The authors, a large Turkish-American collaboration, found that mutations in a gene, WDR62, are associated with severe brain malformations in 9 patients. But what's interesting is that it doesn't cause any particular malformation.

If you have two faulty copies of this gene, your brain won't be normal, but what goes wrong varies widely amongst different people. Although the 9 cases had some features in common, such as microcephaly (small head and brain), in other respects they differed greatly.

As the authors put it, mutations in WDR62 cause

a wide spectrum of severe cerebral cortical malformations including microcephaly, pachygyria with cortical thickening as well as hypoplasia of the corpus callosum. Some patients... had evidence of additional abnormalities including lissencephaly, schizencephaly, polymicrogyria and, in one instance, cerebellar hypoplasia, all traits traditionally regarded as distinct entities.

These are distinct entities, in the sense that you can have any one of them, without having the others. And they are different brain changes. What the authors mean is that everyone assumed that, because they're  different, they must have different genetic causes. They've just shown that this is wrong.

So what is WDR62 "for"? Experiments in mice showed it to be involved in the migration of new neurons from their origin to their final location in the brain. So it's "for" correct neuronal placement, although how it works remains unclear.

WDR62 ought to remind us that there's a long and winding road from gene to phenotype, and that the same gene can, when mutated, cause very different symptoms. This is especially interesting in the light of recent evidence showing that the same mutations can cause a range of behavioural disorders from autism to ADHD to schizophrenia.

Bilgüvar K, et al (2010). Whole-exome sequencing identifies recessive WDR62 mutations in severe brain malformations. Nature, 467 (7312), 207-10 PMID: 20729831

Potential Personal Genomics

A while ago I wrote about how new findings in genetics could herald a new kind of "eugenics", based not around selective breeding to ensure that "bad" genes aren't passed on, but rather based on using fetal genetic testing to choose which variants enter the gene pool in the first place.

I said-

In the near future, we might be able to routinely sequence the genome of any unborn child shortly after conception

But I didn't realize that this may be really very near indeed. Two recent reports have shown that it's possible to sequence fetal DNA from a maternal blood sample. In one case it was used to diagnose a 35 week fetus with a genetic deletion on chromosome 12 seemingly associated with autism, developmental delay and shortness.

In this case it was inherited from the father (which is why they decided to test for it), but this approach could equally be used to screen for the de novo mutations that account for much disease, as I discussed in the last post.

This is big. Currently, the main way to get fetal DNA is through amniocentesis, i.e. inserting a needle into the womb. It's a substantial and not entirely safe medical procedure. A blood sample would be an order of magnitude cheaper and safer, but most of all it would be something you could do at home.

No longer would you need to go to a hospital and discuss everything with a doctor. You could take some blood, send it off anonymously to a sequencing company, and get the results in an email. It would take it out of the hands of professionals and open up a space for individual choice.

The cost of whole-genome sequencing has been falling exponentially and many think it will fall below the $1000 mark within a few years. Combine that with fetal DNA testing and we might see moderately well-off parents able to sequence fetal DNA within the next decade.


When this happens I think the personal genomics industry will suddenly become extremely "hot". At the moment you can sequence your own DNA for a few thousand $ if you want. The results may be interesting but they're of little obvious use. Whatever your genes are, you're stuck with them.

But as soon as we're talking about potential human genomes, it'll kick things up a notch. Media interest and political controversy is sure to follow. Personally I think it'll the debate will begin in earnest when we start seeing selective abortions on the basis of genes for "normal" variants rather than "disease" genes.

It's one thing to not want a child with blindness, or a high risk of leukaemia. But as a society I don't think we're ready for not wanting a child because they're predicted to be a B student rather than an A student, or brunette rather than blonde. At some point soon, though, we'll have to decide what we think about that.

Peters D, Chu T, Yatsenko SA, Hendrix N, Hogge WA, Surti U, Bunce K, Dunkel M, Shaw P & Rajkovic A (2011). Noninvasive prenatal diagnosis of a fetal microdeletion syndrome. The New England journal of medicine, 365 (19), 1847-8 PMID: 22070496

Srebniak M, Boter M, Oudesluijs G, Joosten M, Govaerts L, Van Opstal D, & Galjaard RJ (2011). Application of SNP array for rapid prenatal diagnosis: implementation, genetic counselling and diagnostic flow. European journal of human genetics : EJHG, 19 (12), 1230-7 PMID: 21694736

The Google of Negative Results

A new online resource has been launched which offers us the chance to find out what isn't happening in science.

BioNOT is a free searchable database of negative findings in biology and medicine.

Text mining approaches to the scientific literature have become increasingly popular as a way of helping researchers to make sense of a growing number of papers. But they've tended to focus on positive findings and skim over negative ones. In this sense they're following in the tradition of scientists themselves, unfortunately.

It's also hard to search for negative findings on PubMed, because if you type in, say, vaccines NOT associated with autism in the hopes of finding papers showing that vaccines don't cause autism, it will think you are trying to search for "vaccines" and don't want to see any papers mentioning the words "associated with autism". So you end up with 160,000 hits about vaccines with no reference to autism at all. There are ways around this but it's surprisingly tricky.

BioNOT uses text mining to mine null findings from a large database which includes everything you can find on PubMed and also a large number of full text articles (some behind paywalls).

Authors Agarwal et al of Wisconsin say that this will help to map out the "incidentalome" (a brilliant word I'd never heard before) for a given disease or trait i.e. the regions of the genome that turned out not to be associated with it. It should work for anything, though, not just genes.

However the BioNOT system isn't perfect. The authors note that it is rather over-enthusiastic in finding negative sentences.

A quick try on the system bears this out. I searched for 5 HTTLPR, the claimed "happiness gene". This revealed many papers finding no link between the gene and various things. But it also threw up false positives (how ironic), such as:

young rhesus monkeys were split into two groups... those having, or not, the short variant of the 5 -HTTLPR polymorphism

This is just telling us about the methods of a study. It's not a null finding, but it set the BioNOT alarm bells ringing, presumably because it contained the word "not".

So BioNOT is only a first step, but it's an important one.


Agarwal S, Yu H, & Kohane I (2011). BioNOT: A searchable database of biomedical negated sentences. BMC bioinformatics, 12 (1) PMID: 22032181

The Teen Happiness Gene?

Whether you were happy with life as a teenager could be down to a certain gene, says a new study.

In a large study of American adolescents, the AddHealth project, teens who carried the long form of the 5HTTLPR locus were more likely to say they were satisfied or very satisified with their lives (at age 18 to 26). People with two long variants were the most cheerful, with short/long carriers in the middle and short/short being the least so.

The effect was significant controlling for ethnicity (p=0.013), however looking at the data shows that this effect was largely driven by the unhappy teens who reported being "Dissatisfied" or "Neither" on the 5 point scale of life satisfaction - but there were only a small number of these, because the great majority said they were "Satisfied" or "Very Satisfied". Still, there you go.

Incidentally, Neuroskeptic readers may remember AddHealth because of its role in the "black women are ugly" race row from earlier this year.

This study is the latest in a long, long line of attempts to correlate 5HTTLPR with happiness, depression, stress and so on. A few months ago I discussed the history of this busy little gene and covered a meta-analysis of no fewer than 54 papers which claimed that there was indeed a link, with the short allele increasing the risk of depression in response to stressful events.

However many studies failed to find one, and worryingly the three largest studies were all negative which is a classic tell-tale sign of publication bias - maybe people were only bothering to publish smaller studies if they did find a link and hence were "exciting findings". This is quite possible because so many researchers collect DNA as part of psychology studies these days. When the 5HTTLPR story got big (about 5 years ago) I know a lot of people decided to jump on the bandwagon by looking at it in the context of their old data.

Personally I have no idea whether 5HTTLPR is associated with anything. I used to think it probably did, but now I'm just confusion. There have been so many studies and so much inconsistency that it's very hard to know. What worries me is that I'm not sure whether we'll ever get a consensus. We've already had a gigantic study (over 80,000 people) showing no link and many meta-analyses coming to different conclusions.

What will it take to settle the issue? An even bigger study? Would 200,000 people do it? A million? I don't know.

De Neve JE (2011). Functional polymorphism (5-HTTLPR) in the serotonin transporter gene is associated with subjective well-being: evidence from a US nationally representative sample. Journal of human genetics, 56 (6), 456-9 PMID: 21562513

Men, Women and Spatial Intelligence

Do men and women differ in their cognitive capacities? It's been a popular topic of conversation since as far back as we have records of what people were talking about.

While it's now (almost) generally accepted that men and women are at most only very slightly different in average IQ, there are still a couple of lines of evidence in favor of a gender difference.

First, there's the idea that men are more variable in their intelligence, so there are more very smart men, and also more very stupid ones. This averages out so the mean is the same.

Second, there's the theory that men are on average better at some things, notably "spatial" stuff involving the ability to mentally process shapes, patterns and images, while women are better at social, emotional and perhaps verbal tasks. Again, this averages out overall.

According to proponents, these differences explain why men continue to dominate the upper echelons of things like mathematics, physics, and chess. These all tap spatial processing and since men are more variable, there'll be more extremely high achievers - Nobel Prizes, grandmasters. (There are also presumably more men who are rubbish at these things, but we don't notice them.)

The male spatial advantage has been reported in many parts of the world, but is it "innate", something to do with the male brain? A new PNAS study says - probably not, it's to do with culture. But I'm not convinced.

The authors went to India and studied two tribes, the Khasi and the Karbi. Both live right next to other in the hills of Northeastern India and genetically, they're closely related. Culturally though, the Karbi are patrilineal - property and status is passed down from father to son, with women owning no land of their own. The Khasi are matrilineal, with men forbidden to own land. Moreover, Khasi women also get just as much education as the men, while Karbi ones get much less.

The authors took about 1200 people from 8 villages - 4 per culture - and got them to do a jigsaw puzzle. The quicker you do it, the better your spatial ability. Here were the results. I added the gender-stereotypical colours.

In the patrilineal group, women did substantially worse on average (remember that more time means worse). In the matrilineal society, they performed as well as men. Well, a tiny bit worse, but it wasn't significant. Differences in education explained some of the effect, but only a small part of it.

OK.

This was a large study, and the results are statistically very strong. However, there's a curious result that the authors don't discuss in the paper - the matrilineal group just did much better overall. Looking at the men, they were 10 seconds faster in the matrilineal culture. That's nearly as big as the gender difference in the patrilineal group (15 seconds)!

The individual variability was also much higher in the patrilineal society, for both genders.

Now, maybe, this is a real effect. Maybe being in a patrilineal society makes everyone less spatially aware, not just women; that seems a bit of a stretch, though.

There's also the problem that this study essentially only has two datapoints. One society is matrilineal and has low gender difference in visuospatial processing. One is patrilineal and has a high difference. But that's just not enough data to conclude that there's a correlation between the two things, let alone a causal relationship; you would need to study lots of societies to do that.

Personally, I have no idea what drives the difference, but this study is a reminder of how difficult the question is.

Hoffman M, Gneezy U, List JA (2011). Nurture affects gender differences in spatial abilities. Proceedings of the National Academy of Sciences of the United States of America PMID: 21876159

New Mutations - New Eugenics?

True or false: you inherit your genes from your parents.


Mostly true, but not quite. In theory, you do indeed get half of your DNA from your mother and half from your father; but in practice, there's sometimes a third parent as well, random chance. Genes don't always get transmitted as they should: mutations occur.

As a result, it's not true that "genetic" always implies "inherited". A disease, for example, could be entirely genetic, and almost never inherited. Down's syndrome is the textbook example, but it's something of a special case and until recently, it was widely assumed that most disease risk genes were inherited.

Yet recent evidence suggests that many cases of neurological and psychiatric disorders are caused by uninherited, de novo mutation events. Here are two papers from the last few weeks about schizophrenia(1,2) - but the story looks similar for autism, intellectual disabilities, some forms of epilepsy, ADHD, and others. Indeed they're often the same mutations.

Biologically, a given mutation is what it is, whether it's de novo or inherited. But on a social and a psychological level, I think there are crucial differences, and in particular I think that if it turns out that de novo mutations are important in disease, we're going to see attempts to take these variants out of circulation - far more so than in the case of the very same genes, were they inherited.

The old eugenics movement was based on the idea that if we stop people with bad genes from breeding - by sterilization, voluntary or otherwise, say - we'll be able to eliminate diseases and other undesirable traits. This idea is now generally regarded as extremely unethical, but many of its opponents have shared with the eugenicists the belief that it could work.

But if de novo mutations are what cause the majority of disease, then this approach would be pointless. Sterilizing certain people, or encouraging the healthy ones to have more children, would never be able to eliminate the 'bad genes' because new ones are being created every generation, pretty much at random.

So the de novo paradigm ought to be welcomed by opponents of eugenics. It wasn't just morally wrong - it was biologically misguided too.

But hang on. This is the 21st century. We have in vitro fertilization (IVF), and you can analyze the genes of an IVF embryo before you decide to make it into a child. In the near future, we might be able to routinely sequence the genome of any unborn child shortly after conception.

From there, it would be a small step to allowing parents to decide not to have children with de novo mutations.

This would be, in its effects, a form of eugenics - in the sense that it would produce the effect that the old eugenicists wanted. No more 'bad' variants, or not nearly as many. Opinions will differ as to whether it's morally different. But I would have said that politically, it's a lot more likely to happen.

I can't see forced sterilization returning any time soon. But if you were expecting a baby and you knew that it was not just carrying your and your partner's DNA, but had also suffered a mutation - might you not want to avoid that?

Psychologically, it matters that it did not inherit the variants. It would be a big step to decide that your child should not inherit part of your own DNA. Of course, some variants are obviously harmful, like one that raises the risk of cancer, and I can't see how anyone would want to pass those on. But think about the grey areas - a variant for social anxiety, mild autistic symptoms, obesity, a personality trait.

You might well feel that carrying that variant is what makes you, you; and so it would be natural for your child to have it. You might decide that if it was good enough for you (and all your ancestors), it's good enough for your children. You might well resent the very idea that it's a 'bad gene' at all, as an attack on your own self-worth.

But none of that applies if it's a de novo mutation. Indeed, quite the opposite - all those same considerations would probably lead you to want your children to carry as close as possible to a carbon copy of your DNA, with no random changes. It was good enough for you.

My point is that I think there will be much more support for the idea of genetic screening or other action against de novo variants than against inherited ones. More people will want it, it will be more socially acceptable, and used more widely. I'm not saying this would be a good or a bad thing, just making a prediction. In the future, diseases and traits that are primarily caused by de novo mutations will increasingly selected against.