Machine learning analysis of Autism phenotype data supports a four-dimensional continuum with three overlapping subtypes
- 1Ulster University
Abstract
Autism Spectrum Disorder (ASD) is a heterogeneous neurodevelopmental condition defined by differences in social communication and restricted, repetitive behaviours. As diagnostic criteria have broadened, ASD is now recognised across a wider range of individuals, raising key questions about its structure: does ASD have discrete sub-types, or is it better conceptualised as a continuous, possibly multidimensional, condition?
We aim to explore whether a multidimensional continuum model more accurately captures the variability within ASD. We analysed a large SPARK phenotypic dataset of medical history and diagnostic surveys (background history, SCO, RBS-R; n=36,710 individuals). We apply and compare two traditional statistical approaches, Factor Analysis and Gaussian Mixture Models, with a modern machine learning technique, the Variational Autoencoder (VAE).
VAEs reconstructed unseen test data with [~]4-fold better accuracy than Factor Analysis, and [~]8-fold better accuracy than Gaussian Mixture Models. We identified four stable latent factors across 100 independently trained VAEs. These four dimensions provide an individual behavioural profile that can be visualized using radar-plots, offering a compact way to compare profiles at the person level. Through further analysis, we found evidence for 3 overlapping clusters or subtypes of ASD identified within the 4D latent space. This work aims to inform new ways of modelling ASD using a VAE that will be able to discern between a continuum or a clustered output and that go beyond binary diagnosis, instead reflecting the complex range of trait profiles, with implications for personalised diagnosis and intervention.
Lay SummaryAutism varies considerably between individuals, making it difficult to fully describe using a single diagnostic category. Using machine learning on a large dataset of Autism survey data, we identified four continuous dimensions of autistic characteristics and three overlapping subtypes, highlighting that individuals can share characteristics across groups rather than fitting into distinct categories. These findings may help provide a more individualised understanding of autism and its diverse characteristics.
Related
- Weakly supervised attention-based learning supports breast cancer detection on deep ultraviolet-excitation fluorescence images
- Beyond Binary Diagnosis: A Genetic Algorithm Optimized Hybrid Machine Learning Model for PCOS Phenotype Prediction
- Machine learning–derived intracellular protein–based prognostic model reflecting a senescence-associated secretory phenotype-driven immunosuppressive microenvironment in lung adenocarcinoma
- GRASS-MIL: graph-based representation and discovery of phenotype-associated spatial structures with multiple instance learning
- A phenotype-molecule transfer learning model: Clinical analysis and mechanistic insights of Yinqiao Qingre Tablet in wind-heat syndrome common cold intervention.
BibTeX
@misc{quigley2026machine,
title = {{Machine learning analysis of Autism phenotype data supports a four-dimensional continuum with three overlapping subtypes}},
author = {H. Quigley and B. Gardiner and L. McDaid and C. O'Donnell},
howpublished = {medRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.27.26361561},
url = {https://www.medrxiv.org/content/10.64898/2026.08.27.26361561v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Machine learning analysis of Autism phenotype data supports a four-dimensional continuum with three overlapping subtypes
AU - H. Quigley
AU - B. Gardiner
AU - L. McDaid
AU - C. O'Donnell
PY - 2026
DA - 2026/08/27
PB - medRxiv
DO - 10.64898/2026.08.27.26361561
UR - https://www.medrxiv.org/content/10.64898/2026.08.27.26361561v1
AB - Autism Spectrum Disorder (ASD) is a heterogeneous neurodevelopmental condition defined by differences in social communication and restricted, repetitive behaviours. As diagnostic criteria have broadened, ASD is now recognised across a wider range of individuals, raising key questions about its structure: does ASD have discrete sub-types, or is it better conceptualised as a continuous, possibly multidimensional, condition? We aim to explore whether a multidimensional continuum model more accurately captures the variability within ASD. We analysed a large SPARK phenotypic dataset of medical history and diagnostic surveys (background history, SCO, RBS-R; n=36,710 individuals). We apply and compare two traditional statistical approaches, Factor Analysis and Gaussian Mixture Models, with a modern machine learning technique, the Variational Autoencoder (VAE). VAEs reconstructed unseen test data with [~]4-fold better accuracy than Factor Analysis, and [~]8-fold better accuracy than Gaussian Mixture Models. We identified four stable latent factors across 100 independently trained VAEs. These four dimensions provide an individual behavioural profile that can be visualized using radar-plots, offering a compact way to compare profiles at the person level. Through further analysis, we found evidence for 3 overlapping clusters or subtypes of ASD identified within the 4D latent space. This work aims to inform new ways of modelling ASD using a VAE that will be able to discern between a continuum or a clustered output and that go beyond binary diagnosis, instead reflecting the complex range of trait profiles, with implications for personalised diagnosis and intervention. Lay SummaryAutism varies considerably between individuals, making it difficult to fully describe using a single diagnostic category. Using machine learning on a large dataset of Autism survey data, we identified four continuous dimensions of autistic characteristics and three overlapping subtypes, highlighting that individuals can share characteristics across groups rather than fitting into distinct categories. These findings may help provide a more individualised understanding of autism and its diverse characteristics.
ER -