{"id":40667,"date":"2013-10-10T10:54:15","date_gmt":"2013-10-10T14:54:15","guid":{"rendered":"http:\/\/1boringoldman.com\/?p=40667"},"modified":"2013-10-10T13:15:55","modified_gmt":"2013-10-10T17:15:55","slug":"dont-know-what-we-think-we-know","status":"publish","type":"post","link":"https:\/\/1boringoldman.com\/index.php\/2013\/10\/10\/dont-know-what-we-think-we-know\/","title":{"rendered":"don’t know what we think we know…"},"content":{"rendered":"\n
In the same old banners…<\/a>, I was talking about two markers in psychiatry at the dawn of this new century that I didn’t understand. After psychiatry announced in 1980 that it was off and running to find some kind of biomedical legitimacy, there were twenty years of preoccupation with neurotransmitters and the medications that altered them in one way or another. But with the coming of a new century, the DSM-5 Task Force announced that it was going to put psychiatric diagnosis on a biological footing [A Research Agenda for DSM-V<\/a>] and the Director of the NIMH proposed that psychiatry was really Clinical Neuroscience, and he mapped the coming decades [Psychiatry as a Clinical Neuroscience Discipline<\/a>]. Both of those assertions mystified me because they weren’t backed up by anything yet solid I could see, remaining in the realm of speculation. It seemed a risky bet. <\/div>\n

Looking back, I can now see that they based their predictions on the promise of some new technologies that they felt sure would be confirming – the cracking of the genome and non-invasive functional neuroimaging. We’ve long known that many mental illness run in families<\/em> [see too big a hurry…<\/a>]. So with modern genetic sequencing techniques, they felt it was only a matter of time before the genetic bases of some mental illnesses would be handed to them on a platter by the herd of researchers looking for the answers. Likewise, we all know that the brain is the seat of matters mental. Being able to study the brain in action with the fMRIs would surely yield a cornucopia of confirmations. So the bet wasn’t so risky after all. All they had to do was wait. It was in the bag. And with luck, the neurochemists might come up with a biomarker or two that sealed the deal. No more turtle<\/em> science. The hares<\/em> had arrived! <\/p>\n

The DSM-5 Project was a colossal bust. They built it, but nothing came, and they had to jettison their grand plan and eke out some controversial and trivial changes to even justify a new edition. One of their big disappointments was in the area of genomics. In spite of the runs-in-families<\/em> thing, the genetic studies came up with slim-to-none<\/em> pickings. All they could say was "it’s complicated." Dr. Insel had dubbed the current era, the Decade of Discovery<\/font><\/strong>. He was right about that! But what they discovered was that they had been very naive about genetics. It really is complicated. And now we have this study that confirms [but doesn’t explain] the complexity: <\/div>\n
\n
No Genetic Influence for Childhood Behavior Problems From DNA Analysis<\/a><\/div>\n
by Maciej Trzaskowski, Philip S. Dale, and Robert Plomin <\/div>\n
Journal of the American Academy of Child and Adolescent Psychiatry<\/font><\/strong>. 2013 52[10]:1048-1056.<\/div>\n

<\/p>\n

OBJECTIVE<\/font><\/strong><\/u>: Twin studies of behavior problems in childhood point to substantial genetic influence. It is now possible to estimate genetic influence using DNA alone in samples of unrelated individuals, not relying on family-based designs such as twins. A linear mixed model, which incorporates DNA microarray data, has confirmed twin results by showing substantial genetic influence for diverse traits in adults. Here we present direct comparisons between twin and DNA heritability estimates for childhood behavior problems as rated by parents, teachers, and children themselves.<\/div>\n
METHOD<\/font><\/strong><\/u>: Behavior problem data from 2,500 UK-representative 12-year-old twin pairs were used in twin analyses; DNA analyses were based on 1 member of the twin pair with genotype data for 1.7 million DNA markers. Diverse behavior problems were assessed, including autistic, depressive, and hyperactive symptoms. Genetic influence from DNA was estimated using genome-wide complex trait analysis [GCTA], and the twin estimates of heritability were based on standard twin model fitting.<\/div>\n
RESULTS<\/font><\/strong><\/u>: Behavior problems in childhood-whether rated by parents, teachers, or children themselves-show no significant genetic influence using GCTA, even though twin study estimates of heritability are substantial in the same sample, and even though both GCTA and twin study estimates of genetic influence are substantial for cognitive and anthropometric traits.<\/div>\n
CONCLUSIONS<\/font><\/strong><\/u>: We suggest that this new type of "missing heritability," that is, the gap between GCTA and twin study estimates for behavior problems in childhood, is due to nonadditive genetic influence, which will make it more difficult to identify genes responsible for heritability.<\/div>\n<\/blockquote>\n
Actually holding this paper in my hand, reading and rereading it did not bring a detailed understanding of their methodology. The best I can come up with is the broadest of strokes. Fortunately, we have an expert who weighed in – Neuroskeptic <\/font><\/strong>– our Neuroscience watcher extraordinaire: <\/div>\n
\n
Child Behaviour: Not In Their Genes?<\/a> <\/div>\n
Discover Blogs<\/font><\/strong> <\/div>\n
By Neuroskeptic <\/div>\n
October 8, 2013 <\/div>\n

A paper just published reports that there are: No Genetic Influence for Childhood Behavior Problems From DNA Analysis. This is pretty big. Using a powerful approach called GCTA, King’s College London researchers Maciej Trzaskowski and colleagues found no evidence that genetics can explain differences in children’s behavioural and conduct difficulties. <\/p>\n

First some background. ‘Missing heritability‘ refers to the fact that genetics has mostly failed to find common genetic variants that are associated with ‘complex traits’ like personality, mental disorer and intelligence. This is surprising because these traits are largely heritable – meaning that they run in families, and that identical twins [with all their DNA in common] tend to be more similar than non-identical ones [with only half]. But if they’re heritable then, by definition, there must be genes behind that. But with a few minor exceptions, over a decade of studies drew blanks. Hence the heritability is missing in our DNA, unaccounted for. <\/p>\n

Yet geneticists finally struck gold – or seemed to – with a new technique called genome-wide complex trait analysis [GCTA]. Instead of looking at each variant individually, GCTA quantifies how genetically similar any two people are as a whole. GCTA has shown that the more genetically similar people are, the more similar they tend to be in terms of complex traits. Hooray – the missing heritability is… well, it’s still missing, but at least we know it’s out there, in small pieces scattered across the genome. <\/p>\n

But this good news only applies to some complex traits, according to Trzaskowski et al. It doesn’t hold for child behaviour. In the TEDS sample of British twins, the authors conducted a simultaneous twin and GCTA study. During childhood these twins were assessed for IQ, height, weight, and a range of ‘behaviour problems’ including symptoms of autism, hyperactivity, psychopathy, conduct disorder and more. <\/p>\n

The results of the twin study said that all of the traits were moderately heritable [roughly 0.5 on the scale of 0 to 1]. But while GCTA confirmed a large genetic influence on the intelligence traits, height and weight, it found no genetic influence on the behaviour measures: The difference between the two sets of traits is stark: <\/p>\n

<\/p>\n

The same picture was seen whether it was parents [shown here], teachers, or the children themselves rating their behavioural symptoms. So, some of the missing heritabilities have been located [if not pinned down], but others are more missing than ever. Trzaskowski call it a “new kind of missing heritability” but I wonder if the name “mystery heritability” is more appropriate in relation to these behavioural problems. What’s going on? <\/p>\n

The authors say that GCTA can only detect additive genetic effects, and can’t detect nonlinear interactions between genes. But the trouble is, the twin data just aren’t consistent with the idea that nonadditive effects explain any of these traits. However, Trzaskowski then suggest that the results make sense if we assume that twin studies substantially overestimate the heritability of behaviour problems [but not intelligence]. <\/p>\n

This might be because behaviour problems are just questionnaire ratings of traits, which are essentially subjective and prone to bias [e.g. by parents who assume that their identical twins ‘must be identical’ and rate them as such], while intelligence tests, although not perfect, are less easily fudged.This account does leave room for some genuinely heritability, but only a little. <\/p>\n<\/blockquote>\n

Mere mortals like you and I can’t vet a study like this. The numbers are too big; the indices too numerous and too complex; and the analyses are for the initiated few. All we can go on are the credentials of the scientists, the sources of the data, the opinions of experts, and intuition. In all four areas, this study feels pretty important to me. And admittedly, it looks at behavioral traits in kids, not major psychiatric syndromes, but what it says [as Neuroskeptic puts it] is still pretty big<\/font><\/strong><\/em>. <\/div>\n