Autism Subtypes Unveiled Through Data Analysis

Largest autism study identifies distinct groups, linking traits to genetic factors for tailored support.

A new study is changing how we understand autism by identifying distinct subtypes. By linking shared traits to specific genetic variations, scientists are paving the way for personalized support and earlier interventions for individuals with autism.

Study Highlights Four Autism Subtypes

Researchers at the Flatiron Institute’s Center for Computational Biology (CCB) and their collaborators utilized data from SPARK, the largest autism study to date, to analyze phenotypic and genotypic data from over 5,000 participants with autism, ages 4–18. The study, published in Nature Genetics, successfully identified four distinct groups based on shared traits and linked them to biological processes associated with specific genetic variants.

According to the CDC, about 1 in 36 children has been identified with autism spectrum disorder (ASD) (CDC, 2023).

“A clinically grounded, data-driven subtyping of autism would really help kids get the support they need early on,” says study co-lead author Natalie Sauerwald, a CCB associate research scientist. She added that knowing a person’s subtype can help caregivers access appropriate resources, especially regarding co-occurring conditions like ADHD or anxiety.

Person-Centered Approach Key to Discovery

The study employed a “person-centered” approach, focusing on the full spectrum of traits an individual might exhibit rather than a single trait like IQ. Olga Troyanskaya, senior research scientist and deputy director for genomics at CCB, noted that this strategy was crucial for discovering clinically relevant autism classes and deciphering the underlying biology.

“Our study takes a ‘person-centered’ approach, in which we focus on the full spectrum of traits that an individual might exhibit rather than just one trait, like IQ. This approach was key to our discovery of these clinically relevant autism classes and to deciphering the biology that underlies them.”

—Olga Troyanskaya, senior research scientist and deputy director for genomics at CCB

Kelsey Martin, executive vice president of autism and neuroscience at the Simons Foundation, emphasized the power of leveraging machine learning approaches to analyze the extensive phenotypic and genotypic data available in SPARK.

SPARK: A Landmark Autism Research Initiative

SPARK, supported by the Simons Foundation Autism Research Initiative (SFARI), aims to improve the lives of people with autism by identifying its causes and supporting research that informs more effective therapies, treatments, services, and support. To date, SPARK has engaged over 150,000 individuals with autism and more than 200,000 of their family members.

“I think [SPARK] is the only cohort that has this combination of extensive phenotypic data as well as genetic data,” said Sauerwald.

Identifying Distinct Classes

The researchers classified SPARK participants into four main groups:

• Social and Behavioral Challenges: Individuals in this group, around 37% of participants, have co-occurring traits such as ADHD, anxiety disorders, depression, and mood dysregulation, alongside restricted or repetitive behaviors and communication challenges. They tend to hit developmental milestones at the same rate as their neurotypical peers.
• Mixed ASD with Developmental Delay: Representing approximately 19% of the participants, this group experiences developmental delays but typically does not have the same issues with anxiety, depression, mood dysregulation, or disruptive behaviors.
• Moderate Challenges: This group, comprising roughly 34% of participants, shows challenges similar to the Social and Behavioral group but to a lesser degree and without developmental delays.
• Broadly Affected: The smallest group, accounting for about 10% of participants, is characterized by widespread challenges, including restricted and repetitive behaviors, social communication difficulties, developmental delays, mood dysregulation, anxiety, and depression.

Troyanskaya notes that these classes aren’t definitive but a significant starting point, stating, “This doesn’t mean that there’s necessarily only four classes. I think what this demonstrates is that there are at least four classes. But having the four, which are clinically and biologically relevant, is significant.”

Genetic Pathways Unveiled

Studying the genetics within each class revealed that genetic variants affected biological processes in very distinct ways, with little overlap in the impacted pathways between the classes. Each autism subtype had its own biological signature.

“There was little to no overlap in the impacted pathways between the classes,” says Litman. “And what was even more interesting is that while the impacted pathways – things like neuronal action potentials or chromatin organization – were all previously implicated in autism, each one was largely associated with a different class.”

The team also discovered that the timing of gene activation differed by class. Impacted genes in the Social and Behavioral Challenges class were mostly active after birth, while those in the ASD with Developmental Delays class were primarily active prenatally.

Future Research Directions

Researchers aim to incorporate more data types, including the non-coding portion of the genome, to further refine their understanding of autism subtypes. Sauerwald notes, “The more data, the more discovery. We know there’s a lot of contribution from the non-coding genome in autism, but we haven’t been able to study it yet in the context of these classes. So a big next step is going to be adding in this other 98 percent.”