A letter was published this month in Nature Genetics by Gaugier et al. which has received notable attention. While many of us may be familiar with rare variants of genes that show strong association to autism, such as CNTNAP2 or FMR1, this new study instead focused on the possibility that common genetic variants, i.e., gene polymorphisms that occur in the general population, could affect autism risk via an additive effect.
The authors report that:
“[autism's] narrow-sense heritability is ~52.4%, with most due to common variation, and rare de novo mutations contribute substantially to individual liability, yet their contribution to variance in liability, 2.6%, is modest compared to that for heritable variation” (p. 1).
Now, there’s a lot of Population Genetics terms in that statement so here’s the basic translation: Common point mutations, aka “SNPs,” may confer higher autism risk when additive. This appears to account for roughly 50% of potential genetic risk for autism. Meanwhile, even though rare de novo genetic variants confer considerable risk on a case by case basis, over all, they don’t account for the bulk of genetic risk in the larger autism population (2.6%).
Liability in Population Genetics, image borrowed from here.
Keep in mind that this is just a model which is attempting to estimate how much common genetic variants could affect autism risk. Also keep in mind that they may confer “risk” and do not necessarily “cause” autism. Even rare variants which confer considerable liability are not always associated with autism. Many of the syndromic forms of autism are good examples of this, and while they may more frequently show autism symptomotology, it is a rare genetic syndrome that approaches 100% autism comorbidity.
Right now I’m working on a publication involving a core set of rare genetic variants in autism, studying functional overlap. So when I read this article, I was naturally curious to find out whether some common genetic variants had already been identified, as the authors never made mention of this. So I wrote to Dr. Kathryn Roeder, one of the corresponding authors for the piece, asking exactly this. In a very kind fashion, she responded to my inquiry, stating that there were few, if any, studies addressing this issue, primarily due to such low power. (In order to study common genetic variants, usually very large numbers of individuals are needed in order to supply adequate statistical power, which can be a real challenge in the research of specific conditions like autism.) But she also said she was hopeful that this would start to change in future. I’m likewise very hopeful for the same, especially with the little nudge in interest given by their recent study. I will be chomping at the bit to find out if the functions of common variants overlap those of rare ones. Only time will tell!
