Autism Spectrum Disorder (ASD) is a group of heterogeneous disorders associated with hundreds of susceptibility genes or genetic risk variants of varying frequency and inheritance patterns. The prevalence of ASD in the U.S. is 1 in 44 children, with two thirds of the ASD cases without intellectual disability (ID). Previous genetic studies were often conducted in samples with both ASD with and without ID participants, revealing different genetic architecture of ASD with and without ID: ASD with ID is more associated with de novo variants while ASD without ID has a higher heritability. In addition to ID, other comorbidities including sleep disturbances and metabolic dysfunction were reported in ASD. To further explore the genetic architecture of ASD, especially ASD without ID, we performed an integrative analysis to evaluate the joint contributions of common and rare single-nucleotide variants, small insertions and deletions, and rare copy number variants (CNVs) to the risk of ASD in a new Penn-based collection of ASD without ID probands and family members (Autism Spectrum Program of Excellence (ASPE) collection). We showed in ASPE families, both common and rare variants contributed to ASD and quantitative autistic traits measured by the Social Responsiveness Scale. We also explored a spectrum of phenotypes associated with NRXN1: a known ASD associated trans-synaptic gene. The analysis of the Penn Medicine Biobank identified NRXN1 CNV carriers and found preliminary evidence that NRXN1 CNVs are associated with decreased cholesterol level, which could help explain metabolic dysfunction in ASD. Finally, since mating pattern determines the genetic structure of a population, we investigated assortative mating in ASD with and without ID families in two ASD collections: Simons Foundation Powering Autism Research for Knowledge (SPARK) and the Simons Simplex Collection (SSC). We observed similar degree of positive assortative mating in SPARK and SSC families with and without ID. This work highlighted the joint contributions of different types of genetic variants in multiple genes to the risk of ASD and presented potential gene-gene interaction models that could be studied in future experiments.