That autism can be caused by genetic variations is no longer a mystery. But that some of these variations are found in the general population, in people who have not had an autism diagnosis, is already more troubling. This is shown by a vast study published in the journal Nature Medicine, led by Professor Thomas Bourgeron of the Human Genetics and Cognitive Functions Laboratory at the Institut Pasteur.
“We have worked on several cohorts, including Spark in the US and UK Biobank in the UK. These databases include, in particular, the genetic information of 13,000 autistic people and that of nearly 200,000 people from the general population. We compared these two groups and surprisingly found that certain genetic variations very strongly associated with autism were also present in the general population and went unnoticed”, explains the researcher.
The scientists were particularly interested in a type of genetic variation called “loss of function”, which strongly impacts the function of the gene by swapping part of the protein it encodes. These variations, relatively rare, are present in 4% of autistic people but also, more surprisingly, in 1% of the general population. The geneticists then wanted to verify whether these genetic modifications had an impact on non-autistic individuals.
The finding is clear: “The non-autistic individuals who make up this 1% carrier of rare genetic variations have on average a significantly lower IQ than the rest of the population, a lower level of education, a lower salary and greater material deprivation. However, these people, while exposed to the same types of genetic variation as the 4% with autism, have not been diagnosed and may not experience all of the same challenges.
The study goes even further. It highlights the complex genetic architecture of autism with two extremes. In some cases, a single genetic variation is enough to have a diagnosis of autism. In others, it is the accumulation of thousands of genetic variations that will increase the likelihood of being autistic.
Among the earliest so-called “monogenic” forms of autism, the study shows that about a dozen genes are never found mutated in the general population. In this case, it is a genetic variation that appears de novo in the child while it is not present in its parents. These variants have a strong effect. The presence of one of them is often associated with forms of autism with intellectual disability and very pronounced language disorders.
But they can accumulate in children and contribute to the development of autism. In this case, it will rather be an Asperger-type autism, without intellectual deficit therefore. Of course, between the two, all combinations are possible to reflect the great diversity of autism spectrum disorders.
But here again, Thomas Bourgeron’s team looked at the general population to identify what the effects of the accumulation of these polygenic variants could be. This time people have a higher IQ and more years of education. On the other hand, they have rather average salaries and greater material deprivation. In other words, the accumulation of low-effect genetic variations (polygenic variants) is rather associated with higher IQ and higher education, but also, paradoxically, with socio-economic difficulties.
