Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition characterized by chronic impairment in social communication and the presence of restricted or repetitive behaviors, which has been increasingly prevalent across the world-wide \cite{Ousley14} \cite{Lord20}. High prevalence rates are consistently reported in large-scale epidemiological studies, underscoring the need for trustworthy and scalable computational tools for early detection and diagnosis \cite{Shaw25}. Genomic studies demonstrated that ASD is highly heritable and genetic, with rare variants and polygenic risk contributing to a convergence of neurodevelopmental pathways. Advances in gene expression profiling technologies have facilitated the generation of high-dimensional transcriptomic datasets and promise to provide an opportunity for machine learning in autism classification and biomarker discovery \cite{Qiu22} \cite{Gogate24}.

Autism Spectrum Disorder (ASD) represents a complex condition affecting brain development that causes ongoing challenges with social communication and includes limited or repetitive behavioral patterns. This disorder has become more commonly recognized around the globe. Large population studies consistently show high occurrence rates, highlighting the urgent need for reliable and expandable computer-based methods to identify and diagnose ASD early. Research into genetics has revealed that ASD runs strongly in families and has genetic foundations, where uncommon genetic variations and multiple genetic risk factors work together through shared brain development processes. Recent improvements in technology that measures gene activity have enabled scientists to create detailed datasets containing vast amounts of genetic information. This progress opens up possibilities for using artificial intelligence to classify autism cases and identify biological markers of the condition.