A study published online in Cell reports the latest investigation of de novo germline mutation by whole genome sequencing in autism patients. This study provides a global view of the landscape of mutability and its influence on genetic diversity and susceptibility in autism, and its implications on other human diseases. The work was a collaborative effort led by international teams comprised of the University of California, San Diego, BGI, and other institutes. The results are expected to shed new light on a deeper understanding of the mechanisms underlying genome evolution and human diseases.

Mutation plays an important role in human diseases, such as Autism Spectrum Disorders (ASDs). Many cases for ASDs are caused by de novo mutations that are not inherited, but arise spontaneously in the ovum, sperm or fertilized egg. Epidemiologists have reported a higher risk of autism in children with older fathers, but so far there has been few biological evidence to support this theory. To comprehensively explore the genetics of ASDs, it is vital to understand the mutational process and how the de novo germline mutation impacts ASDs.

In this study, researchers applied whole genome sequencing (WGS) approach to characterize patterns of de novo germline mutations (DNMs). A total of 581 DNMs were identified from ten monozygotic twins that suffered from ASDs by comparing with their unaffected parents. To better understand the paternal age effects on mutation rate, the twins were separated into two groups, one with younger fathers (<30 years old), and the other with older fathers (>40 years old). The results showed that paternal age accounted for a substantial portion of variability in mutation that happened in offspring, while maternal age has no significant effect.

Mutation is a random process. However, researchers in this study found DNMs displayed a remarkably non-random positioning in the genome and spaced more closely than their expectation. More importantly, the distribution of de novo germline mutation can be explained by characteristics of the genome. Clusters of new mutations could be explained by allelic gene conversion or compound mutation. Clustering on larger scales could be explained by mutation-rate variation. The regional mutation rates are subject to a combination of influences, including DNase hypersensitivity, GC content, nucleosome occupancy, recombination rate, simple repeats, the trinucleotide sequence surrounding the site, among others.

Researchers further examined the landscape of mutability throughout the genome, including hotspots with highly elevated mutability, and warm spots with moderately increased mutability. They found some strong evidences to support that hypermutability is a characteristic of disease genes. Intriguingly, they found the mostly highly mutable sequences in the genome are the most highly conserved. This finding has not been reported in the previous studies.

Another interesting result was that the genes impacted by DNMs in twins demonstrated a significant association with autism in other independent projects. These findings suggest that regional hypermutability is a significant factor shaping patterns in genetic variation and disease risk in humans.

Yujian Shi, Project Manager of BGI, said, "The study opens a new way for pedigree studies on neurological diseases and rare diseases. The novel approach and results here will help to massively analyze lineage or sporadic autism population. We found there was a significant relationship between human diseases and individual genetic variation model shaped by DNMs derived regional hypermutability. Furthermore, the discovery of a large number of novel autism susceptibility genes will lay a solid foundation for the early diagnosis and treatment of autism."