Systems-level multi-omics dissection of syndromic and idiopathic autism reveals distinct regulatory architectures, candidate molecular signatures, and potential therapeutic targets.

Biological systems operate as self-organizing information networks in which genetic, epigenetic, and regulatory interactions collectively determine functional outcomes. Autism encompasses a heterogeneous set of neurodevelopmental conditions, including syndromic and idiopathic subtypes. Despite extensive gene discovery efforts, how these autism subtypes differ in their underlying organization of biological information remains poorly understood. Here, we apply an integrative systems-level, multi-omics framework to compare syndromic and idiopathic autism as distinct regulatory systems. High-confidence autism risk genes were curated from the SFARI and AutismKB databases, and analyzed using functional enrichment, protein-protein interaction network modeling, graph-theoretic hub identification, brain-region, cell-type-specific transcriptomic validation, experimentally supported miRNA regulatory network reconstruction, and deep learning-based drugtarget interaction analysis. Our analyses reveal clear differences in network organization between autism subtypes. Idiopathic autism is predominantly associated with synaptic signaling, ion channel activity, and transcriptional modulation, with hub genes KAT2B and AR enriched in basal ganglia-associated regions and astrocytes. In contrast, syndromic autism shows enrichment for transcriptional regulation, chromatin remodeling, and dense miRNA-mediated control, with hub genes CHD3 and CSNK2A1 preferentially expressed in cerebellar and cortical regions, as well as inhibitory neurons. Notably, master regulatory miRNAs differ completely between subtypes, indicating distinct post-transcriptional regulatory strategies. Deep learning-based screening further identifies subtype-specific therapeutic candidates with predicted central nervous system accessibility. Together, these findings demonstrate that syndromic and idiopathic autism differ in how regulatory information is structured and propagated across molecular networks, providing a systems-level perspective on autism heterogeneity and a general framework for analyzing biological information organization in complex systems.