Brain samples of autism patients have revealed widespread similarities in the regulation of gene expression – a stark contrast from the diversity of changes in their DNA. This is the first study to see clear differences in the DNA and protein packaging of autistic brains compared with normal brains, says Shyam Prabhakar, a computational biologist at the A*STAR Genome Institute of Singapore, who led the analysis. “I’m excited about collaborating with experimental scientists and drug companies to advance this research.”

But the lack of uniformity in the genetic code of autistic patients has stymied drug development. “A hundred or more genes are associated with autism, which is an awful lot of complexity when designing a drug to treat the majority of patients.” This diversity has caused some to question whether autism can be categorized as a single disease.

So Prabhakar teamed up with Daniel Geschwind in the United States and Jonathan Mill in the United Kingdom to see if they could find biological convergence – if not in the DNA itself, then in how it was expressed.

They studied around 250 brain samples from patients diagnosed with autism and matched controls, each group examined different molecules in the brain samples.

Prabhakar’s team looked at modifications to the histone proteins around which DNA is wrapped. These proteins can alter a gene’s accessibility and thus its expression. Specifically, the team analyzed histone tags known as H3K27ac, found at regions of DNA that activate gene expression.

The researchers were thrilled to see thousands of differences between the autistic and normal tissue. “Histone acetylation is systematically altered in the autistic brain,” says Prabhakar. Even cases where autism was just one aspect of a broader set of symptoms, such as chromosomal duplication, agreed with the pattern of histone tagging observed in cases of unknown origin.

The genes linked to these histone modifications belonged to several categories. There were “the usual suspects” – synaptic, ion channel, and immune system genes – and some “new categories” not previously linked with autism – certain signaling–protein ligand and receptor genes, as well as some involved in early structural development.

Prabhakar hopes to follow these leads toward potential drug targets. Moreover, he says, the findings demonstrate the effectiveness of histone acetylome–wide association studies for understanding other diseases. “It’s a gold mine – a huge untapped area of research that should be applied to a whole bunch of diseases.”