livinglegend/Shutterstock.com

The genetic roots of autism are as diverse as the disorder itself. Some mutations are inherited, whereas others arise spontaneously — sort of. Results from two new studies serve as a reminder that duplications and deletions of some chromosomal regions, which can increase the risk of autism, are not as spontaneous as they may seem.

It’s true these deletions and duplications often arise anew in people with autism. But they only occur because of stretches of DNA inherited from the unaffected parents that flank the region. The flanking segments stick to each other, causing the DNA to break and re-form at different points.

In one new study, published 19 October in Nature Genetics, researchers focused on the DNA flanking 15q13.3, an approximately 2-megabase stretch of DNA. Deletions in this region are associated with epilepsy, intellectual disability and autism.

Looking at the genomes of 2,235 people from around the world, the team found five distinct versions of these flanking sequences that differ by as many as 1,000 nucleotides. Some versions may be more likely to lead to 15q13.3 deletion than others.

The researchers looked at these flanking regions in sequences from 86 nonhuman primates, including chimpanzees, orangutans, gorillas and bonobos. They also looked in detail at sequences from a Neanderthal and a Denisovan — early hominids that diverged from humans about 700,000 years ago. Their findings suggest that regions flanking 15q13.3 are uniquely human and probably emerged during human evolution.

Work presented at the 2014 American Society of Human Genetics Annual Meeting in San Diego suggests that the regions surrounding another autism-linked region, 16p11.2, also arose during human evolution. In this case, too, some people may be more prone to the duplications or deletions than others.

In the case of 15q13.3, one particular gene within the flanking sequence, dubbed GOLGA8, may be the key driver. GOLGA8 is part of a 15-gene family that arose and expanded across the genome over the last 20 million years of primate evolution. Sequences from six people, one chimpanzee and two gorillas suggest that the 15q13.3 segment breaks off right next to a palindromic GOLGA8 repeat, causing the segment to be deleted.

Taken together, the studies support the intriguing hypothesis that the same genetic complexity that makes us human also renders us more susceptible to changes that drive neurological diversity. A less prosaic theory is that these repeats are simply side effects of an expanded and complex genome.

Whichever theory you prefer, we may need to expand our idea of inheritance from our immediate ancestors to our species as a whole.