February 03, 2010

Genome-wide associate studies and cognitive performance

This ties in with what I've written before, namely, that complex phenotypes don't have a set of population-common genetic causes, but are caused by different combinations of "commodity" alleles.

Just as you can drink water from a cup, or a glass, or a bucket, or a spoon, or your hands, i.e., you can achieve the same functional effect by very different means, so you can achieve the same phenotypic effect (e.g., intelligence) by very different genetic means.

Using a different analogy, the effect of a "cross" can be achieved both like this:


**0**
**0**
00000
**0**
**0**


and like this:


00*00
00*00
*****
00*00
00*00


You wouldn't have a clue that these ASCII characters both create a "cross" if you limited yourself to testing whether a bit of information was "0" or "*". It is in their arrangement that the pattern emerges.

So it is with many complex human phenotypes. As humans, we are machines with billions of independent variables (alleles) which can produce the same (or similar) outcomes in a manifold number of ways.

European Journal of Human Genetics doi: 10.1038/ejhg.2010.2

Common genetic variation and performance on standardized cognitive tests

Elizabeth T Cirulli et al.

Abstract

One surprising feature of the recently completed waves of genome-wide association studies is the limited impact of common genetic variation in individually detectable polymorphisms on many human traits. This has been particularly pronounced for studies on psychiatric conditions, which have failed to produce clear, replicable associations for common variants. One popular explanation for these negative findings is that many of these traits may be genetically heterogeneous, leading to the idea that relevant endophenotypes may be more genetically tractable. Aspects of cognition may be the most important endophenotypes for psychiatric conditions such as schizophrenia, leading many researchers to pursue large-scale studies on the genetic contributors of cognitive performance in the normal population as a surrogate for aspects of liability to disease. Here, we perform a genome-wide association study with two tests of executive function, Digit Symbol and Stroop Color-Word, in 1086 healthy volunteers and with an expanded cognitive battery in 514 of these volunteers. We show that, consistent with published studies of the psychiatric conditions themselves, no single common variant has a large effect (explaining >4–8% of the population variation) on the performance of healthy individuals on standardized cognitive tests. Given that these are important endophenotypes, our work is consistent with the idea that identifying rare genetic causes of psychiatric conditions may be more important for future research than identifying genetically homogenous endophenotypes.

Link

8 comments:

Maju said...

True but you also have epigenetics, which is not just environmentally caused but also sometimes inheritable.

So while I don't mean to question that genes matter (they do of course, though maybe not as much as we used to believe) I support the notion that much is environmental rather than just genetically pre-determined. Not just you can get a cross with two different sets (or many more probably) but also you can modify the cross at a second stage (epigenetics), activating, neutralizing or regulating it in different ways.

This is good for diversity and diversity is good for adaptability, so that's why it exists after all.

marnie said...

I've heard various scientists discuss over the last few years, the guess that schizophrenia involves the widely distributed duplication or deletion of genetic information across the genome. There is no one type of failure. People with this terrible disease present a broad combination of genetic errors.

Here is the link to the discussion:

http://www.npr.org/templates/
story/story.php?storyId=89163694

(I'll post the text if, for some reason, the link doesn't come up.)

There is a lot of redundancy built into the brain, it seems.

On a related note, it is interesting that men present a broader variation of intelligence than women. To be honest, I think much of this is cultural, but some of it I would suspect is genetic.

I'm not sure if I have this right, but the Y on the XY of the twenty third chromosome is "empty" of a lot of expressed gene information. So the 23 XY chromosome lacks redundancy, compared to the 23 XX? Is that right?

If so, it does make me wonder. Men are then "under constrained", gene expression wise, compared to women?

Another thought on genetic redundancy is that it allows us to adapt to our environment. A long forgotten gene can suddenly find a new use, when called upon.

Overall, there seems to be a genetic struggle between redundancy and leaning toward more intelligence.

Interesting, hey?

Maju said...

I'm not sure if I have this right, but the Y on the XY of the twenty third chromosome is "empty" of a lot of expressed gene information. So the 23 XY chromosome lacks redundancy, compared to the 23 XX? Is that right?.

Essentially yes, I understand. However the issue seems more complex. AFAIK:

1. In females one of the two X chromosomes is randomly switched off in each cell at some developmental stage.

2. There seems to be paternally and maternally switched genes in diverse regions of the whole genome (similar as with female X but in diverse segments of other chromosomes and non-random - or seems so).

Overall, there seems to be a genetic struggle between redundancy and leaning toward more intelligence.

I don't understand why would be this way. I'm not even sure there is any clear tendency towards more intelligence at the genetic level (just improved nutrition, health and education). Actually I suspect we are somewhat dumber that our Paleolithic ancestors because since Neolithic we don't have to face the usual survival challenges so much and since civilization intelligence is a secondarily important trait: obedience, tameness is much more important - and we all know that wolves are smarter than dogs, right?

marnie said...

"There seems to be paternally and maternally switched genes in diverse regions of the whole genome (similar as with female X but in diverse segments of other chromosomes and non-random - or seems so)."

Really? Is there a paper on that? Very interesting.

"I'm not even sure there is any clear tendency towards more intelligence at the genetic level (just improved nutrition, health and education). Actually I suspect we are somewhat dumber that our Paleolithic ancestors because since Neolithic we don't have to face the usual survival challenges so much and since civilization intelligence is a secondarily important trait: obedience, tameness is much more important - and we all know that wolves are smarter than dogs, right?"

Ha!

Well, I'm not necessarily looking at Neolithic vs. Paleolithic. All in all, that is a pretty short timescale. And yes, the Lascaux paintings always catch me, whenever I look at them, as if I was looking at some groovy artistic hipster painting his or her heart out.

I don't think we've fully adapted to the supposed current "requirements" of "modern" society. Most of us seem to be chafing at the bit of tameness.

But I think we've long had to struggle with our tendancies toward individualism versus necessities of cooperation.

Speaking of dogs (and wolves), they do seem to have greater capacity for cooperation than humans. (I'm thinking of several wonderful books on the topic: "Never Cry Wolf" by Farley Mowat and "The Secret Life of Dogs" by Elizabeth Marshall Thomas.)

I had the privilege of living in Ghana, West Africa, when I was a kid. In all this discussion about human genetics and genetic distance, I often hear of some "great leap forward" in the last 100 thousand years, or something to that affect. Yet, the Ghanaians, in small ways, seem to have some abilities which my European relatives do not. One small example I can think of is tremendous rhythmic capacity in music. Another thing I can think of is vocal quality. The vocal quality, particularly the capacity to project a rich base, seems to be a broad speacial quality in the voices of Ghanaian men. I'll confine my discussion to that!

These are subtle population differences.

But what we are really saying is that there is great inertia against changes in human intelligence, probably due to the fact that so many genes are involved.

A demain!

Maju said...

I commented briefly on this issue of paternally/maternally activated gene batches at this post at Leherensuge. The original paper is this one (open access): Japanese researchers found that fatherless mice, in spite of being smaller, also lived some 20% more. The issue of paternally/maternally activated genes is dealt with as an already known issue but I don't know too much on the matter myself (follow the footnotes of the paper if you want to dig on it).

I often hear of some "great leap forward" in the last 100 thousand years.

Well, there are several theories in this regard. What you mention here by the date, seems to refer to the so called "modern human behaviour", which is nothing but direct evidence of symbolic behaviour, whose first fossil evidence is found since roughly those dates (maybe as old as 130 Kya in Palestine). But no fossil evidence would ever tell us about music and dance for example. We have always to relativize the fossil evidence: it gives a minimal age but tell us nothing about the real age, probably much older. Not just we may well be missing other fossil evidence but a most of what hunter-gatherers use leaves no (or very weak at best) traces.

The true biological leap happened in the 600-200 Kya period, when hominins reached brain capabilities close to ours and even surpassing them in some cases after this last date, leading to Neanderthals and our species too. Since then it has remained roughly stable, though we can't say much about "wiring", only size.

Anyhow, since H. erectus, our genus has got very notable brain capabilities (c. 900 cubic centimeters on average). While some individuals surpass this size (about 1050 c.c.) since c. 1.2 million years ago, others do not and the average seems stable until the leap mentioned above.

But early H. sapiens (and Neanderthals too) had already big brains some 190 and 160 Kya, when the first fossils are found. So, even if the direct evidence for symbolic behavior (shell ornaments specifically) does not appear till later, I think they were perfectly capable of it and surely expressed that themselves in ways that just have not left fossil remains such as dance, music and other perishable forms.

Ponto said...

I don't know if there is a paper on it but it has been noted for ages. The Barr body is essentially a switched off X.

Human females are essentially what are called mosaics. Not every cell in their body is identical as not the same X is switch off in every cell.

Women who carry the color blindness gene usually have more color vision but if you were to test small sections of their eye you'll find some are color blind and others have typical color acuity.

It is interesting.

The Neolithic actually created a lot of issues. The biggest ones were disease. Diseases crossing over from animals, zoonoses, and diseases due to being located in one spot for a long time. That would have put the Neolithic farmers under stress to either survive these diseases or try to limit their extent by hygiene and cultural habits.

Judging from the effects on populations of the Americas and Oceania of the diseases brought into their proximity by European colonialists in recent history, the Paleolithic peoples were probably bowled over like tenpins by these Neolithic farmer diseases. When North America was colonized most of the east coast was underpopulated by Amerindians, many having died previously by diseases introduced by Europeans in South, Central and southern North America that had moved north up the east coast prior to North America been colonized.

Maju said...

Ponto: America and Australia were very much isolated world regions. We do not know of any major bacterial shock in the Old World, not even in New Guinea (but it happened in Andaman, another hyper-isolated region). Germs have been flowing around the Old World with relative ease all the time. European foragers were not isolated from West Asian (or even other Asian) peoples at any long period, nor were South Asian nor African ones.

marnie said...

"The true biological leap happened in the 600-200 Kya period, when hominins reached brain capabilities close to ours"

That seems right to me. Seems to fit more closely with what you see in people.

"We have always to relativize the fossil evidence: it gives a minimal age but tell us nothing about the real age, probably much older."

Astute observation.

"Human females are essentially what are called mosaics."

Ponto, yes, this makes sense, at the macro level.

I'm trying to suggest that human intelligence is a constrained optimization problem:

http://en.wikipedia.org/wiki/Constraint_optimization

The number of "constraints" for male and female intelligence, statistically speaking, may be slightly different. I'm actually not sure, I'm just wondering about this. (Based on the number of genes that contribute to intelligence.)

"The Neolithic actually created a lot of issues. The biggest ones were disease. Diseases crossing over from animals, zoonoses, and diseases due to being located in one spot for a long time. That would have put the Neolithic farmers under stress to either survive these diseases or try to limit their extent by hygiene and cultural habits."

There are many documented tragic examples of these diseases. The effect of smallpox introduced by Europeans to the Americas was clearly devastating for indigenous peoples.

Generally, I would say that coping with all of the various transmissible diseases has placed a tremendous limit on human capacity.

The prevalence of disease in different populations likely plays a much greater role in human capacity than variations in genetic intelligence.