Artificial Intelligence Vs. Autism: How A New Machine-Learning Approach Can Be Useful In Autism Genes Identification

By Kristine Walker, Parent Herald August 02, 12:30 pm
Close
Drone race: Nasa's AI-backed drone takes on human-controlled one

Artificial intelligence (AI) has been a valuable asset in the field of medicine, particularly its potential to unlock the possibilities for personalized medical treatments. Due to the ubiquity of AI in precision medicine, a team of Princeton University and Simons Foundation researchers developed a machine-learning technique to identify the genes that causing autism spectrum disorder (ASD).

Based on the study published in the journal Nature Neuroscience, scientists developed a "complementary" artificial intelligence-driven approach through the use of a functional map of the brain (brain-specific gene network) to offer a genome-wide prediction of autism risk genes. The researchers also created a "user-friendly, interactive web portal" that will allow biomedical experts to access and investigate the findings of the research.

"Our work is significant because geneticists can use our predictions to direct future sequencing studies, enabling much faster and cheaper discovery of autism genes," Princeton University associate research scholar and study lead author Arjun Krishnan said, as per EurekAlert!. "Researchers can use our predictions to prioritize and interpret results of whole-genome sequencing studies of ASD."

Princeton University's computer science and genomics professor and senior study author Olga Troyanskaya also stressed that the newly developed artificial intelligence-driven approach for autism gene identification can determine ASD-associated genes even if they have not been previously linked to autism. But in order to achieve such finding, a brain-specific gene network produced by integrating thousands of genomic datasets should be used.

Meanwhile, the genes that play a vital role in autism are reportedly emerging. According to The Scientist, the mutation in a gene called chromodomain helicase DNA binding protein 8 (CHD8) is currently one of several dozens of recognized genetic subtypes of autism but psychologist Raphael Bernier believes there could be more.

Due to the discovery of molecular, cellular and anatomical indicators of autism such as genetic mutations, experts has proven that autism is not a single biological disorder. As a matter of fact, autism types and causes are as innumerable as its symptoms, not to mention the various types of successful treatments.

In other autism-related news, dogs can reportedly help children with autism. University of Lincoln veterinary behavioral medicine professor Daniel Mills said that they found an important connection between "parenting stress of the child's main caregiver and their attachment to the family dog," UPI noted.

In the study published in the Journal of Veterinary Behavior, experts found that parents of a child with autism who are also dog owners have experienced less stress and anxiety. In addition, the study also highlighted the positive impact of pet ownership in the long-term improvement to the lives of families with kids with autism.

What do you think about the latest AI-driven approach in autism genes identification? Sound off below and follow Parent Herald for more news and updates.

© 2017 ParentHerald.com All rights reserved. Do not reproduce without permission.

Sign up for our Newsletter

Real Time Analytics