A new genetic study on autism spectrum disorders (ASD) conducted by Cold Spring Harbor Laboratory (CSHL) and the University of Pennsylvania has shown that damaging de novo mutations—that is, those that occur randomly and are not inherited genetically from the parents—are “significantly and convincingly related to measures of impaired motor skills.”
The study, published in the Proceedings of the National Academy of Sciences, titled “Damaging de novo mutations diminish motor skills in children on the autism spectrum,” used DNA sequence and deep phenotypic data from the Simons Simplex Collection, a core project and resource of the Simons Foundation Autism Research Initiative (SFARI). The set is comprised of data from 2,760 families that have a single child affected by ASD and was collected over five years from 12 universities across the U.S. and Canada.
Molecular biologist Michael Wigler, Ph.D., a professor at CSHL, conducted the study along with CSHL professor Ivan Iossifov, Ph.D., who provided the overall molecular framework, and first author Andreas Buja, Ph.D., a statistician from The University of Pennsylvania. Wigler said the study is part of an incremental investigation of autism that is “pushing a particular perspective on autism that needs to be amplified.”
A de novo mutation is caused because “DNA is intrinsically prone to mutate,” Wigler said. “It spontaneously decays. [That’s why] all living things have evolved extremely elaborate repair mechanisms. We could not exist without this.” Every once in a while, DNA escapes repair, which, in the case of de novo mutations, causes autism to occur in “a brief window” in the germ line of the parents, he said. The older the parent, the more likely it is to happen, and there is “no constraint” on the type of damage that can occur this way.
It is already known that it is common for children who have autism to have decreased motor skill levels and lower IQs, Wigler said. Currently, children are evaluated according to social and behavioral deficits, not motor skills. The study revealed that there is a clear connection between genetic mutation and motor skills, and it is more sensitive than IQ.
Wigler said the study was done in incremental parts. The first showed that when children had known de novo mutations, their IQs were much lower “and we extended that to motor skills, so that there’s a correlation between having a damaging mutation and having decreased motor skills,” he said. The second phase of the research showed that motor skills is the more sensitive indicator of that damage than is IQ. Part three found there does not appear to be a correlation between the damaging mutations and behavioral defects, which “creates something of a puzzle that needs explanation,” he said.
The paper does not provide an explanation with proof, but it does provide a variety of hypotheses, Wigler said. The one he favors proposes that genetic autism is milder because those carrying it were able to function well enough to have children, those who could not would have died without passing the gene on.
According to Wigler, the study was possible because of the nature of the collection, which focused on simplex autism, or families with only one child with autism. Most collections are multiplex, with two or more children in the family with the disease, he said.
Catherine Lord, Ph.D., an author on the paper, and a professor at Columbia University, has been involved in the development of standardized diagnostic instruments for ASD. She said the study is helpful because there isn’t much research on motor skills and autism, in part because because there are fewer measures available for motor skills, while IQ measurements are more standardized. Scientists still do not know exactly how the motor disabilities in ASD children relate to IQ, she said.
Lord doesn’t feel the study is likely to immediately change how autism is evaluated, but she hopes it might be used to tell families with an individual child with autism what patterns to expect in the child’s development, so that treatment can be more focused.
Lord was instrumental in collecting the data and is pleased with the standards that were used, how it was entered and curated, and with the complex math that went behind drawing information from the data, while controlling for many variables. “It really is an amazing dataset,” she said, when compared with the National Institutes of Health’s National Database for Autism Research, available by grant, that does not provide the exact diagnosis of the child and doesn’t have standards for how the data are collected.
Professor David Skuse, M.D., an expert in ASD and head of Behavioural and Brain Sciences at the Institute of Child Health in London, said the study is “the most carefully done work of its type” that he has seen. The U.K. has a similar dataset in size and genetic anomalies, he said. “We will look at our data in light of these findings. It offers a very compelling analysis.”
Skuse said he is impressed with the thought that went into the interpretations of data, the genetic anomalies and how they were be classified. The 15-question survey used in the study to collect information from parents provided valuable information and could be built upon in the future.
In the study, researchers said they used their finding to propose a combined classification of phenotypic severity: mild (little impairment of either), moderate (impairment mainly to motor skills), and severe (impairment of both IQ and motor skills).
“The way [the researchers] presented the relationship between these classes of genetic anomaly and phenotypic data has not been seen before. It clearly requires further consideration,” he said. “The study is exemplar and sets up a template for future studies.”
Click here to access the Clinical OMICs digital issue.