Big data approaches to decomposing heterogeneity across the autism spectrum

Abstract

Autism is a diagnostic label based on behavior. While the diagnostic criteria attempts to maximize clinical consensus, it also masks a wide degree of heterogeneity between and within individuals at multiple levels of analysis. Understanding this multi-level heterogeneity is of high clinical and translational importance. Here we present organizing principles to frame the work examining multi-level heterogeneity in autism. Theoretical concepts such as 'spectrum' or 'autisms' reflect non-mutually exclusive explanations regarding continuous/dimensional or categorical/qualitative variation between and within individuals. However, common practices of small sample size studies and case-control models are suboptimal for tackling heterogeneity. Big data is an important ingredient for furthering our understanding heterogeneity in autism. In addition to being 'feature-rich', big data should be both 'broad' (i.e. large sample size) and 'deep' (i.e. multiple levels of data collected on the same individuals). These characteristics help ensure the results from a population are more generalizable and facilitate evaluation of the utility of different models of heterogeneity. A model's utility can be shown by its ability to explain clinically or mechanistically important phenomena, but also by explaining how variability manifests across different levels of analysis. The directionality for explaining variability across levels can be bottom-up or top-down, and should include the importance of development for characterizing change within individuals. While progress can be made with 'supervised' models built upon a priori or theoretically predicted distinctions or dimensions of importance, it will become increasingly important to complement such work with unsupervised data-driven discoveries that leverage unknown and multivariate distinctions within big data. Without a better understanding of how to model heterogeneity between autistic people, progress towards the goal of precision medicine may be limited.. CC-BY-NC-ND 4.0 International license peer-reviewed) is the author/funder. It is made available under aThe copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/278788 doi: bioRxiv preprint first posted online Mar. 9, 2018; 3 Autism occurs in approximately 1-2% of the population 1 and autistic individuals'wellbeing is a major public health issue. In economic terms, the lifetime individual cost of autism is estimated at $2.4 (£1.5) million in the USA and UK and annual population costs are around $268 billion in the USA 2,3 . While interest in and science on autism has been growing rapidly, progress towards translating scientific knowledge into high-impact clinical practice has been less rapid. We are still far from delivering more effective intervention and support, more precise and earlier diagnosis, better understanding and prediction of prognosis and development, and personalization of support and intervention. All of these points are within the scope of stratified psychiatry 4 and precision me...

Reconciling Dimensional and Categorical Models of Autism Heterogeneity: a Brain Connectomics & Behavioral Study

“…First, most previous ASD subtyping studies assumed that each participant belonged to a single (categorical) subtype. By contrast, the term "spectrum" in ASD suggests continuous variation across individuals (18). This is observed at varying degrees across multiple symptom domains (19,20).…”

Intrinsic excitation-inhibition imbalance affects medial prefrontal cortex differently in autistic men versus women

“…While no one theory can fully explain all individuals with an autism diagnosis 4,5 , the E:I imbalance theory may have utility for understanding subtypes of autistic individuals [6][7][8] . Sex/gender may be an important stratifier of relevance for highlighting E:I imbalance subtypes 9,10 .…”

Probing disrupted neurodevelopment in autism using human stem cell‐derived neurons and organoids: An outlook into future diagnostics and drug development

“…It is now clear that ASDs are complex and multifaceted disorders that can be caused by disruptions in multiple biological pathways and, as such, are unlikely to be fully dissected using a "one size fits all" model. 343 There is a need to understand whether and how disruption in various stages of neurodevelopment may lead to a common set of symptoms that are observed in specific subsets of ASD patients in light of patients' genetic background (sub-phenotyping at the cellular level). Therefore, new approaches are needed to systematically identify and characterize the biological convergence points that can serve as biomarkers for specific autistic phenotypes.…”