Autism gene discoveries more than tripled, revealing complex genetic networks

A new analysis of genetic data has more than tripled the number of high-confidence genes linked to autism, offering fresh insights into the diverse pathways shaping the condition and highlighting the need for more inclusive research.

A new preprint has more than tripled the number of high-confidence genes linked to autism, lifting the total to 253 and giving researchers a clearer picture of how different genetic pathways may shape the condition. The work, reported by The Transmitter, builds on a landmark 2022 analysis that identified 72 genes and began separating autism-predominant genes from those tied to broader neurodevelopmental disorders.

The latest analysis drew on exome data from 62,429 autistic people in the Autism Sequencing Consortium and three other large cohorts. According to the report, the expanded dataset strengthened the earlier finding that some autism-linked genes are concentrated in synaptic biology, while others are more closely associated with transcriptional regulation, a process that controls when genes switch on and off. Researchers said the genes could be grouped into six clusters based on overlap with schizophrenia, bipolar disorder, epilepsy and other neurodevelopmental conditions, suggesting that autism genetics is not a single pathway but a network of partially overlapping ones.

The studies also point to meaningful differences in development and outcome. Genes more specific to autism were more often tied to mature excitatory neurons and synapse assembly, while genes associated with a wider range of conditions were more strongly expressed in immature neurons and linked to transcriptional regulation. In the companion analysis, children carrying rare variants in genes connected to multiple conditions were more likely to experience developmental delays and less likely to reach milestones than children carrying variants in autism-specific genes. Mark Daly of the Broad Institute said the groupings were informative but not predictive, cautioning that individual genetic results cannot yet forecast a child’s trajectory.

A third preprint looked at the size of autism risk carried by de novo, or spontaneous, coding variants. Using genetic data from 38,680 autistic people and their parents and non-autistic siblings, the researchers estimated that de novo variants account for 3.4 per cent of the variation in genetic susceptibility to autism. About half of that contribution can be traced to just 15 genes, most of them associated with severe neurodevelopmental conditions. The model also suggested that around 7 per cent of autistic people carry a spontaneous variant predicted to raise the likelihood of the condition, with the strongest effects concentrated in genes under strong evolutionary constraint.

Even with the larger gene list, the authors said many autism-linked genes remain undiscovered and that better data collection will matter as much as sample size. The forecast analysis suggested that roughly a quarter of strongly linked genes have yet to be found, and that discovery could accelerate if studies include more autistic people with intellectual disability. Maria Chahrour of the University of Texas Southwestern Medical Centre, who was not involved in the preprints, told The Transmitter that the result was “a practical, useful finding for how future studies should be designed,” while also arguing for greater ancestral diversity and more attention to gene-environment interactions. UCSF and Broad Institute work cited in related coverage likewise underscores that autism genetics continues to expand rapidly, even as researchers refine which genes matter most and how they shape brain development.

Disclaimer: This content is for informational purposes only and is not intended to be a substitute for professional medical judgment, advice, diagnosis, or treatment.