Cross-regional and transcript-level prioritization of alternative-splicing events in autism cortex

Introduction Selecting splice events for functional studies requires disease-associated RNA measurements to be connected to regulatory context and explicit transcript consequences. We developed an event-level prioritization framework for this task in autism spectrum disorder (ASD). Methods We integrated 34,025 published cortical splicing tests with regional direction, donor-aware sensitivity analyses, external cortical datasets, neural RNA-binding-protein data, and reconstructed coding branches. Results An effect-size-informed rule identified 278 exploratory candidates; no cortical event met transcriptome-wide false-discovery-rate (FDR) significance (minimum FDR = 0.1969). Among direction-evaluable events, 183/267 candidates were concordant in cerebellum, compared with 14,405/26,915 background events (68.5% versus 53.5%; odds ratio [OR] = 1.89). Because 46/48 cerebellar donors also contributed cortex, this comparison measures same-cohort coherence, not independent replication. RBFOX2 overlap showed a model-dependent adjusted association (OR = 1.94), without significant within-gene or binding-permutation support. Separate healthy motor-neuron datasets showed adjusted ORs of 2.16 and 2.13 (both q = 0.0059). Coding reconstruction classified 92/148 coding-overlap events as in-frame changes and distinguished transcript-dependent consequences; a PKM exon exchange preserved protein length while substituting 22 amino acids. External ASD analyses did not establish FDR-corrected replication, and baseline-adjusted neural SRRM4 comparisons were nonsignificant. The framework nominated 18 events under the complete-coding-model scope, including 13 retained under the stricter scope; both branches were observed for 12/18 and 11/18 in separate fetal bulk and single-cell catalogs. Discussion The resulting resource connects each nomination to a defined regulatory or coding rationale and an identifiable transcript structure, providing a focused basis for isoform-resolved experimental studies.