Thursday 5 February 2015

Your IQ in 13 genes (or about 29% of it)

I have already said that I favour the gene hunters over the gene saboteurs. That said, I keep a bleary and cynical eye open when gene hunters find something, lest I be led up the garden path by false positives.

http://drjamesthompson.blogspot.co.uk/2014/09/gene-hunters-and-gene-saboteurs.html

I am not against false positives per se. Psychology specialises in them. By ensuring that most published work is based on small unrepresentative samples, mostly of psychology students, psychology researchers ensure that a steady stream of false findings are fed to the newspapers, who are in dire straights because no-one reads them any more, and they need as much seductive nonsense as they can lay their hands on in order to keep their remaining readers.

And yet, and yet, if one manages to put together a very large sample, and checks the findings against those obtained from other large samples, it ought to be possible to run one’s fingertips across the face of reality. Gene hunters tend to lead the pack when it comes to sample size.

With this in mind it is a rare pleasure to report the findings of a team who have found something. Of course a null result is as important as a positive result. We are all adults here. However, it is good to come across a result in the usual sense of that term, a finding which excites, rather like Raedwald’s  Saxon sword at Sutton Hoo.

Published the day before yesterday, in Molecular Psychiatry, advance online publication, 3 February 2015; doi:10.1038/mp.2014.188,  this multi-author paper merits the breathless categorization of an “Immediate Communication”. In short, they have made a major advance in tracking down some genes which are very likely to be involved in intelligence. Here is the abstract: 

General cognitive function is substantially heritable across the human life course from adolescence to old age. We investigated the genetic contribution to variation in this important, health- and well-being-related trait in middle-aged and older adults. We conducted a meta-analysis of genome-wide association studies of 31 cohorts (N = 53 949) in which the participants had undertaken multiple, diverse cognitive tests. A general cognitive function phenotype was tested for, and created in each cohort by principal component analysis. We report 13 genome-wide significant single-nucleotide polymorphism (SNP) associations in three genomic regions, 6q16.1, 14q12 and 19q13.32 (best SNP and closest gene, respectively: rs10457441, P = 3.93 × 10− 9, MIR2113; rs17522122, P = 2.55 × 10− 8, AKAP6; rs10119, P = 5.67 × 10 − 9, APOE/TOMM40). We report one gene-based significant association with the HMGN1 gene located on chromosome 21 (P=1×10− 6). These genes have previously been associated with neuropsychiatric phenotypes.
Meta-analysis results are consistent with a polygenic model of inheritance. To estimate SNP-based heritability, the genome-wide complex trait analysis procedure was applied to two large cohorts, the Atherosclerosis Risk in Communities Study (N = 6617) and the Health and Retirement Study (N = 5976). The proportion of phenotypic variation accounted for by all genotyped common SNPs was 29% (s.e. = 5%) and 28% (s.e. = 7%), respectively. Using polygenic prediction analysis, ~ 1.2% of the variance in general cognitive function was predicted in the Generation Scotland cohort (N = 5487; P = 1.5 × 10− 17). In hypothesis-driven tests, there was significant association between general cognitive function and four genes previously associated with Alzheimer’s disease: TOMM40, APOE, ABCG1 and MEF2C.

https://drive.google.com/file/d/0B3c4TxciNeJZUjdEVHF4dkVtaTQ/view?usp=sharing

The paper has a particularly good introductory section, which is worth reading in its own right as a starting point in understanding intelligence and ageing.

Participants were individuals from 31 population-based cohorts of European ancestry aged 45 years or older excluding dementia and clinical stroke (including self-reported stroke). The total sample size was 53 949 individuals (N men = 23 030, N women = 30 919).  Intelligence was estimated by extracting a principal components factor from a range of heterogenous mental tasks, and this factor accounted for 34% to 62% of the variance. Therefore, large populations given the same mental tasks might provide even higher g estimates, though the procedure seems to work well with disparate measures.

However, if one moves from the samples of discovery to a sample of replication then only 1.2% of the variance in that new sample can be explained by the patterns detected in the samples of discovery. Nonetheless, to be able to explain so much variance in a human behaviour on the basis of so few genes is intriguing, particularly when these genes are linked in other studies with cognitive ageing.

This study provides further evidence that general cognitive function is heritable and under polygenic control. These findings are consistent with, and add considerably to those from the Cognitive Ageing in Genetics in England and Scotland consortium.

In conclusion, we report the largest meta-analysis of GWAS studies, to date, of fluid general cognitive function in middle and older age. We also report results showing that general cognitive function is heritable and highly polygenic, extending findings of previous studies involving general cognitive function in older individuals. We show genome-wide significant SNP-based associations within three genomic regions 6q16.1 (MIR2113), 14q12 (AKAP6/NPAS3 region) and 19q13.32 (TOMM40/APOE region), and a genome-wide significant gene-based association with the HMGN1 gene located on chromosome 21. The 19q13.32 region has long been associated with AD and more recently was associated with non-pathological cognitive aging; the 6q16.1, 14q12 and HMGN1 regions contain genes associated with development of the
brain, neurological function, psychiatric disease and educational attainment.

Now the race is on to confirm, disconfirm and extend these findings.

9 comments:

  1. "in dire straights": you won't find those at the Independent. I suspect you mean "in dire straits".

    ReplyDelete
    Replies
    1. Yes, already apologized on Twitter to the first person to point that out

      Delete
  2. "if one moves from the samples of discovery to a sample of replication then only 1.2% of the variance in that new sample can be explained by the patterns detected in the samples of discovery": it therefore sounds to this sceptic to be a damp squib. But I compliment them on the honesty of that test. Anyway, given that IQ is approximately normally distributed, shouldn't one expect to find that the genetic causes are many and small?

    ReplyDelete
    Replies
    1. I agree. When reading the part you quoted I too felt like it implies that their findings are not meaningful. Going from 29% of variance explained to 1.2% is not much different than going to 0%.

      Delete
    2. 29% is the variance explained by all SNPs in a GCTA design. Specific, identified variants do not explain 29% in any sample. With a sample large enough, they will be able to explain 29% using just common SNPs.

      Delete
    3. Thank you for your explanation, which I will have to follow up in a later post so as to amend and expand the account I gave. (Silence on this for a while till I do more reading). Also, can you give an indicator of which Anon you are? Helps keep track of contributor's contributions and reply properly

      Delete
  3. What is the iq that was analysed??


    Santoculto

    ReplyDelete
    Replies
    1. Variety of different tests, hence the need to work with just the principal component.

      Delete
    2. You do not think that some Cultural ( contextual) ccomponent could be interesting to analyse as part of intelligence?? I think "observational capacity" to understand the reality is a very important component to live and to adapt in most of different social environment.

      Santoculto

      Delete