published in Nature Genetics. The findings, based on analysis of more than 150,000 people’s genetics, arose from a collaboration of five research groups whose work included comparisons of ASD cohorts with separate cohorts of individuals with developmental delay or schizophrenia.
“We know that many genes, when mutated, contribute to autism,” and this study brought together “multiple types of mutations in a wide array of samples to get a much richer sense of the genes and genetic architecture involved in autism and other neurodevelopmental conditions,” co–senior author Joseph D. Buxbaum, PhD, director of the Seaver Autism Center for Research and Treatment at Mount Sinai and a professor at the Icahn School of Medicine at Mount Sinai, both in New York, said in a prepared statement. “This is significant in that we now have more insights as to the biology of the brain changes that underlie autism and more potential targets for treatment.”
Glen Elliott, PhD, MD, a clinical professor of psychiatry at Stanford (Calif.) University who was not involved in the study, said the paper is important paper for informing clinicians of where the basic research is headed. “We’re still in for a long road” before it bears fruit in terms of therapeutics. The value of studies like these, that investigate which genes are most associated with ASD, is that they may lead toward understanding the pathways in the brain that give rise to certain symptoms of ASD, which can then become therapeutic targets, Dr. Elliott said.
Investigating large cohorts
The researchers analyzed genetic exome sequencing data from 33 ASD cohorts with a total of 63,237 people and then compared these data with another cohort of people with developmental delay and a cohort of people with schizophrenia. The combined ASD cohorts included 15,036 individuals with ASD, 28,522 parents, and 5,492 unaffected siblings. The remaining participants were 5,591 people with ASD and 8,597 matched controls from case control studies.
In the ASD cohorts, the researchers identified 72 genes that were associated with ASD. De novo variants were eight times more likely in cases (4%) than in controls (0.5%). Ten genes occurred at least twice in ASD cases but never occurred in unaffected siblings.
Then the researchers integrated these ASD genetic data with a cohort of 91,605 people that included 31,058 people with developmental delay and their parents. Substantial overlap with gene mutations existed between these two cohorts: 70.1% of the genes related to developmental delay appeared linked to risk for ASD, and 86.6% of genes associated with ASD risk also had associations with developmental delay. Overall, the researchers identified 373 genes strongly associated with ASD and/or developmental delay and 664 genes with a likely association.
“Isolating genes that exert a greater effect on ASD than they do on other developmental delays has remained challenging due to the frequent comorbidity of these phenotypes,” wrote lead author Jack M. Fu, of Massachusetts General Hospital and Harvard Medical School, both in Boston, and colleagues. “Still, an estimated 13.4% of the transmission and de novo association–ASD genes show little evidence for association in the developmental delay cohort.”