Summary: A new computational model pinpoints genes linked to autism and can predict the degree of intellectual disability in people with ASD by analyzing rare mutations in genes implicated in the disorder.
Source: Baylor College of Medicine
Researchers estimate that hundreds of genes may play roles in autism spectrum disorder (ASD), but distinguishing disease-causing mutations from incidental variants has been a persistent challenge.
A study led by investigators at Baylor College of Medicine and published in Science Translational Medicine demonstrates a novel computational strategy that effectively highlights genes most likely involved in ASD. The approach also predicts intellectual disability severity in ASD patients by focusing on rare, de novo missense mutations and leveraging evolutionary data to prioritize variants with functional impact.
Identifying genes that contribute to ASD enables researchers to investigate the biological pathways that underlie the condition, improve risk prediction, and offer better guidance to families about potential outcomes and early interventions.
“ASD is highly complex, and many cases lack an obvious genetic explanation with current knowledge,” said first author Dr. Amanda Koire, who began this work as a graduate student in the Dr. Olivier Lichtarge laboratory and is now a psychiatry research resident at Brigham and Women’s Hospital, Harvard Medical School.
The team emphasizes that there is no single gene responsible for most ASD cases. “The most frequently mutated genes associated with ASD explain only about 2% of cases,” said Olivier Lichtarge, Cullen Chair and professor of molecular and human genetics and related fields at Baylor. “Current thinking is that ASD arises from many gene mutations, each contributing a small effect.”
Because contributory variants are individually rare, identifying them through single-patient analysis is unlikely to succeed. Even population-wide comparisons between affected individuals and unaffected relatives often leave many cases unexplained.
To overcome these limitations, the Baylor team took a different perspective. They incorporated extensive evolutionary information to assess which variants are most likely to alter protein function. Evolutionary data provide a rich record of how changes in protein sequences affect fitness over deep time, offering a powerful filter to flag damaging mutations.
The researchers then applied two additional filters: they concentrated on personal, or de novo, mutations that arise anew in an individual and considered the cumulative impact of these mutations within molecular pathways rather than focusing exclusively on single genes.
Exploring the contribution of de novo missense mutations in ASD
The study focused on missense variants, a common class of mutations that substitute one amino acid for another. Unlike loss-of-function mutations that often inactivate a protein entirely, missense changes can have subtle or severe effects on protein function, making their interpretation more difficult.
“Loss-of-function mutations have been tied to ASD severity, including lower motor skills and IQ, but establishing a large-scale link between missense variants and clinical features has been challenging,” said co-author Dr. Panagiotis Katsonis, assistant professor of molecular and human genetics at Baylor. “People with ASD are more likely to carry de novo missense changes than de novo loss-of-function variants, and our lab’s tools help interpret these prevalent coding variants.”
De novo mutations arise for the first time in an individual and are not inherited from either parent. The researchers aimed to distinguish which de novo missense variants found in ASD probands were more likely to differentiate affected individuals from their unaffected siblings.
A multilayered approach
The team used a multilayered strategy to identify genes and mutations most likely to contribute to ASD. They began by analyzing whole-exome sequences from 2,392 families in the Simons Simplex Collection to catalog de novo missense variants across all protein-coding genes.
Each missense mutation was scored using the Evolutionary Action (EA) equation, a computational method developed in the Lichtarge lab that quantifies the likely effect of an amino acid substitution on protein fitness. EA scores range from 0 to 100, where higher scores indicate a greater predicted fitness impact on the protein.
Among 1,418 de novo missense mutations affecting 1,269 genes in the ASD cases, most genes were altered only once. Given the multigenic nature of ASD, the authors reasoned that damaging mutations could be distributed across components of the same biological pathway in different individuals, rather than consistently hitting the same gene.
Without preselecting candidate genes or pathways, the researchers asked which pathways contained an excess of de novo missense variants with elevated EA scores. Using this cohort-level, pathway-focused analysis, they identified 398 genes across 23 pathways with a significant bias toward higher EA scores than expected by chance.
Noteworthy pathways included axonogenesis—the process of axon formation in neurons—synaptic transmission, and other neurodevelopmental processes. In axonogenesis, in particular, multiple missense variants collectively showed a strong tendency toward high EA scores, indicating impactful changes across that pathway.
“By layering complementary perspectives on the likely functional impact of variants, we were able to pinpoint a set of genes clearly related to ASD,” Lichtarge said. “Many of these pathways are neurologically relevant; some genes were already linked to ASD and others represent novel candidates.”
The team also found a relationship between EA scores in candidate genes and patient IQ. For newly implicated genes, higher EA scores corresponded to an average decrease of about seven IQ points in affected individuals, suggesting that these variants exert measurable biological effects.
“This opens multiple avenues,” said co-author Young Won Kim, a graduate student in Baylor’s Integrative Molecular and Biomedical Sciences program. “We now have new genes to study in ASD and a potential way to inform families about likely outcomes and to prioritize early developmental support, which can greatly influence long-term results.”
Lichtarge added that the approach could extend beyond ASD. “As genomic sequencing becomes more available, applying evolutionary-action-based interpretation to rare variants could clarify the polygenic basis of many complex diseases and improve individualized risk and morbidity estimates.”
Christie Buchovecky (Baylor and Columbia University) and Stephen J. Wilson (Baylor) also contributed to this work.
Funding: This research received support from the National Institutes of Health (grant numbers GM079656-8, DE025181, GM066099, AG061105), the Oskar Fischer Foundation, the National Science Foundation (DBI1356569), the Defense Advanced Research Projects Agency (N66001-15-C-4042), RP160283 – Baylor College of Medicine Comprehensive Cancer Training Program, the Baylor Research Advocates for Student Scientists (BRASS), and the McNair MD/PhD Scholars program.
About this genetics and autism research news
Source: Baylor College of Medicine
Contact: Molly Chiu – Baylor College of Medicine
Image: The image is in the public domain
Original Research: Closed access. “A method to delineate de novo missense variants across pathways prioritizes genes linked to autism” by Amanda Koire et al., Science Translational Medicine
Abstract
A method to delineate de novo missense variants across pathways prioritizes genes linked to autism
Genotype–phenotype relationships shape health and population fitness but are challenging to predict and interpret. This study applies an evolutionary-action method to de novo missense variants in whole-exome sequences from individuals with autism spectrum disorder to identify genes and pathways connected to ASD.
Evolutionary Action predicts the impact of missense variants on protein function by estimating fitness consequences based on phylogenetic distances and substitution probabilities among homologous sequences.
Comparing de novo missense variants in 2,384 individuals with ASD to matched unaffected siblings, the study identified 398 genes across 23 pathways with a bias toward higher EA scores than expected by chance; these pathways included axonogenesis, synaptic transmission, and neurodevelopmental processes.
Predicted fitness impacts of de novo and inherited missense variants in candidate genes correlated with IQ in individuals with ASD, even for genes newly implicated by this analysis. Using the evolutionary-action framework, the authors prioritized missense variants likely to contribute to ASD pathogenesis and demonstrated their phenotypic relevance. This cohort-level approach could be applied to other complex diseases to aggregate and interpret missense variation that contributes to shared phenotypes.
