‘Machine-Learning’ Scans Genomic Database for Autism Clues
In an effort to better analyze a huge genomic database for autism-related genes, a team of engineers have teamed up with autism researchers to develop a “machine learning” program. So far, the method has identified several genes that help scientists better understand the development of autism and other conditions.
The research, supported in part by advocacy and science group Autism Speaks, is led by Stephen Scherer, Ph.D., director of the Autism Speaks genomics program MSSNG (a genomic database). He is also the director of the University of Toronto McLaughlin Centre and The Centre for Applied Genomics at Toronto’s SickKids Hospital.
Scherer has pioneered the use of genomics to provide medically useful information for individuals affected by autism and other developmental disorders.
“Massive genomic databases such as MSSNG are only as good as the tools and innovations in analysis that researchers bring to them,” said Autism Speaks Chief Science Officer Rob Ring.
“Machine-learning is just the type of analysis necessary to realize the full value of the genomic data we’re generating. We love to see this happening here.”
Machine-learning enables a computer to sort through an enormous information database to scan for any significant patterns. In this case, the program is combing through more than 60,000 variations in the human genome to pinpoint any genes that are likely associated with autism and other medical conditions.
The program is currently focusing on a little-explored region of the genome that is responsible for the splicing of messenger RNA. This type of RNA turns DNA into proteins. Proteins, in turn, are involved in all aspects of development and body function.
“This work is groundbreaking because it represents a first serious attempt to decode portions of that 98 percent of the human genome outside the genes that are typically analyzed in genetic disease studies,” said Scherer.
”This is particularly exciting since it is thought these segments may contain much of the missing information that we have been looking for in autism research.”
In a collaborative effort, the team compared mutations in the genomes of children with autism to those in children without autism. Using traditional genetic analysis, they found no differences between the two groups. However, when they used machine-learning to identify mutations that change RNA splicing, surprising patterns emerged.
Overall, they identified 39 previously unrecognized gene changes linked to autism. They also discovered gene changes related to hereditary cancers and spinal muscular atrophy.
The findings are published in the journal Science.
Source: Autism Speaks
Pedersen, T. (2018). ‘Machine-Learning’ Scans Genomic Database for Autism Clues. Psych Central. Retrieved on July 11, 2020, from https://psychcentral.com/news/2015/01/07/machine-learning-scans-genomic-database-for-autism-clues/79574.html