Friday, October 26, 2012

De novo deleterious variants identified via sequencing

Genome sequencing is poised to become a routine clinical tool.
Forbes: In two papers in major scientific journals, researchers today suggested pushing DNA sequencing into more routine use in the clinic, and not just as a research tool.

Dutch researchers are proposing that DNA sequencing replace older forms of genetic tests for diagnosing the cause of severe intellectual disability, the second time in a day that researchers have pushed the emerging technology as a first-choice diagnostic test for severe illness. Those results were published this evening in the New England Journal of Medicine.

“This is the new test for intellectual disability. There is no doubt about it,” says Han Brunner of the Radboud University Nijmegen Medical Center, one of the Dutch study’s authors. “This is a paradigm shift to genome-first medicine for patients who have complex problems that will not be easy to diagnose by conventional strategies.”

Earlier today, a study in Science Translational Medicine proposed that DNA sequencing could become a standard first-choice test for infants in neonatal intensive care units, because a combination of new software and hardware could allow doctors to get results in just 50 hours, answering questions about what is making a baby sick far faster when time is of the essence.
The Dutch researchers found de novo mutations which led to severe cognitive impairment. It is almost certain that there are many more mutations which lead to smaller impairment. This is evidence that mutational load is at least a partial factor in population variation in cognitive ability. See earlier post: Deleterious variants.
NEJM: We evaluated patients with intellectual disability to exclude known causes of the disorder. We then sequenced the coding regions of more than 21,000 genes obtained from 100 patients with an IQ below 50 and their unaffected parents. A data-analysis procedure was developed to identify and classify de novo, autosomal recessive, and X-linked mutations. In addition, we used high-throughput resequencing to confirm new candidate genes in 765 persons with intellectual disability (a confirmation series). All mutations were evaluated by molecular geneticists and clinicians in the context of the patients' clinical presentation.

... We identified 79 de novo mutations in 53 of 100 patients. A total of 10 de novo mutations and 3 X-linked (maternally inherited) mutations that had been previously predicted to compromise the function of known intellectual-disability genes were found in 13 patients. Potentially causative de novo mutations in novel candidate genes were detected in 22 patients. Additional de novo mutations in 3 of these candidate genes were identified in patients with similar phenotypes in the confirmation series, providing support for mutations in these genes as the cause of intellectual disability.

... De novo mutations represent an important cause of intellectual disability; exome sequencing was used as an effective diagnostic strategy for their detection. (Funded by the European Union and others.)
Very rough numbers: suppose 30 de novo mutations per person lead to 1 in 1000 chance of cognitive impairment, so 1/30k of 3E09 bp = 1E05 bp affect cognition in a powerful way. That's 100k loci ... presumably variation at these sites is fixed in the normal population, and (normal) variation in cognitive ability is due to some smaller set of nearby loci.

5 comments:

Richard Seiter said...

One question, what effect would these de novo mutations have in the existing studies attempting to evaluate relative genetic and environmental effects? Since they are essentially random (right?) would they show up as unshared environmental effects (statistically)?

steve hsu said...

De novo mutations with big negative effects on IQ never show up in any heritability studies because individuals with such mutations would be removed (or never recruited in the first place).

De novo mutations with small negative effects on IQ would be lumped into unshared environment, as you say. My guess is that this is not a very big effect.

jeffhsu3 said...

Yes, and though de novo mutations always seemed to be the most parsimonious answer for unshared 'environmental' effects it was often overlooked due to the technological constraints in addressing that question. So genetics really does explain more than just the heritibility estimates, though probably only marginally so. In addition to de novo germline mutations, de novo somatic mutations could explain a little bit more of the variance, but that would require sequencing the relevant tissue which, of course, is a lot harder.

BlackRoseML said...

Well, this study would not be expected to elucidate directly the genetic components that contribute to the substantial genetic influence of intelligence in the general population. For instance, figure four (here: http://webspace.pugetsound.edu/facultypages/cjones/chidev/Paper/Articles/Plomin-IQ.pdf) shows that most cases of severe mental retardation are not heritable or caused by the genetic loci implicated in the normal variation of intelligence in the general population, as mild mental retardation is a phenotype which is caused by possessing a high amount of intelligence-detracting alleles, each minutely affecting intelligence additively. The quantity of one's genetic burden of deleterious variants follows a Gaussian distribution in the population. Thus, mildly mentally retarded people are merely individuals on the tail-end of the normal distribution for the quantity of burdensome alleles; in contrast, the etiology of severe mental retardation tends to be caused by severe congenital genetic disorder, such as aneuploidy or a debilitating de novo mutation, or in utero stress. As a result, since mild mentally retarded children inherit a substantial genetic burden of deleterious alleles from their parents, their siblings would thus be more vulnerable to inherit a similar disadvantageous genetic endowment from their parents, while severe mental retardation is not transmitted from parent to offspring and hence not "heritable".

botti said...

Just regarding psychometric and cognitive research I note the passing of UC Berkeley psychologist Arthur Jensen.

http://m.legacy.com/obituaries/record-bee/obituary.aspx?n=&pid=160674984&referrer=0&preview=True …

Blog Archive

Labels