Excitatory/Inhibitory balance in autism spectrum disorders: Integrating genetic, neurotransmitter and computational perspectives

Computational modeling of excitatory/inhibitory (E/I) balance offers transformative insights into the neurobiological underpinnings of autism spectrum disorder (ASD). In this review, we examined the integration of neurotransmitter dynamics and genetic factors into multiscale computational frameworks to elucidate the mechanisms driving E/I dysregulation in ASD. We explored the pivotal roles of glutamate and GABA, the primary excitatory and inhibitory neurotransmitters, and the modulatory impact of serotonin and dopamine (DA), in shaping neural circuit stability, behavioral outcomes, and ASD core symptoms. Genetic mutations affecting synaptic proteins such as SHANK3, GRIN2A, and GABRB3 were highlighted for their capacity to disturb synaptic scaffolding and glutamatergic and GABAergic signaling, thereby shifting the E/I ratio. Computational approaches, ranging from detailed neuronal simulations to neural mass and spiking network models, captured the heterogeneous manifestations of E/I imbalance and aligned with molecular, neuroimaging, and electrophysiological findings in ASD. We discussed how these models informed individualized diagnostic strategies, enabled prediction of treatment responses, and offered targets for precision medicine. Major challenges included methodological inconsistencies, neurochemical measurement discrepancies, polygenic interactions, and the translation of model predictions into clinical practice. We concluded that the integration of neurotransmitter and genetic data within advanced computational models represents a significant advance toward unraveling ASD pathophysiology, with the promise of developing dynamic, personalized interventions. Ongoing efforts should emphasize longitudinal data, multiomic integration, sex-specific trajectories, and cross-disciplinary collaboration to further the clinical applicability and translational potential of computational E/I balance modeling in autism research.