A new machine-learning approach could predict autism-causing genes in the human genome.
To predict genes that cause autism spectrum disorder (ASD), scientists from Princeton University and Simons Foundation have developed a big data and machine learning approach that maps the human gene network and shows how each gene is functionally related to one another.
"[The method] learns how known autism genes are connected to other genes in a gene network, and then uses these patterns to predict novel ASD genes," Arjun Krishnan, associate researcher scholar at Princeton and first author of the study, said in an interview by ResearchGate.
"The gene network we used represents how genes function together in cellular pathways in the brain, or, intuitively, a molecular-level functional map of the brain."
ASD is a neurodevelopment disorder that has a strong genetic basis. However, only 65 autism genes out of an estimated 400 to 1,000 have been found. Because of ASD's complex nature, genetic sequencing methods are not enough to uncover the genetic basis of autism.
The research team then developed a complementary machine-learning approach using a map of the brain to predict ASD risk genes. After finding hundreds of ASD candidate genes, the scientists used these and the brain network to identify the stages and regions of brain development and the specific cellular functions that might be disrupted in ASD.
The new approach was validated in a large independent case-control sequencing study, and the researchers also created an interactive web portal where biomedical researchers or clinician could access the study results.
Geneticists can use these predictions to direct future sequencing studies to enable faster and cheaper discovery of autism genes, Krishnan said in a press release.
According to the researchers, the study results could help ASD researchers efficiently narrow down the genetic underpinnings of ASD and focus on the identified ASD candidate genes for genetic screens and laboratory experiments.
"We critically need a genetic or molecular test to diagnose ASD and to introduce drugs or other therapeutic interventions as early in brain development as possible based on their genetic makeup," Krishnan said.
The findings of the study were published in the journal Nature Neuroscience.