Access to “big data” is a boon for researchers, fostering collaboration and resource-sharing to accelerate advancements across fields. Yet, disentangling complex datasets has been hindered by methodological limitations, calling for alternative, interdisciplinary approaches to parse manifold multi-directional pathways between clinical features, particularly for highly heterogeneous autism spectrum disorder (ASD). Despite a long history of male-bias in ASD prevalence, no consensus has been reached regarding mechanisms underlying sex-related discrepancies.

Applying a novel network-theory-based approach, we extracted data-driven, clinically-relevant insights from a well-characterized sample (http://sfari.org/simons-simplex-collection) of autistic males (N = 2175, Age = 8.9 ± 3.5 years) and females (N = 334, Age = 9.2 ± 3.7 years). Expert clinical review of exploratory factor analysis (EFA) results yielded factors of interest in sensory, social, and restricted and repetitive behavior domains. To offset inherent confounds of sample imbalance, we identified a comparison subgroup of males (N = 331) matched to females (by age, IQ). We applied data-driven causal discovery analysis (CDA) using Greedy Fast Causal Inference (GFCI) on three groups (all females, all males, matched males). Structural equation modeling (SEM) extracted measures of model-fit and effect sizes for causal relationships between sex, age-at-enrollment, and IQ on EFA-determined factors.

We identified potential targets for intervention at nodes with mediating or indirect effects. For example, in the female and matched male groups, analyses suggest mitigating RRB domain behaviors may lead to downstream reductions in oppositional and self-injurious behaviors.

Our investigation unveiled sex-specific directional relationships that inform our understanding of differing needs and outcomes associated with biological sex in autism and may serve to further development of targeted interventions.